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, Nicolai Stange <nicstange@gmail.com>,
	Theodore Tso <tytso@mit.edu>
Subject: [PATCH 4.4 87/99] ext4: silence UBSAN in ext4_mb_init()
Date: Sun,  5 Jun 2016 14:42:00 -0700	[thread overview]
Message-ID: <20160605213911.491082493@linuxfoundation.org> (raw)
In-Reply-To: <20160605213902.974592018@linuxfoundation.org>

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

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

From: Nicolai Stange <nicstange@gmail.com>

commit 935244cd54b86ca46e69bc6604d2adfb1aec2d42 upstream.

Currently, in ext4_mb_init(), there's a loop like the following:

  do {
    ...
    offset += 1 << (sb->s_blocksize_bits - i);
    i++;
  } while (i <= sb->s_blocksize_bits + 1);

Note that the updated offset is used in the loop's next iteration only.

However, at the last iteration, that is at i == sb->s_blocksize_bits + 1,
the shift count becomes equal to (unsigned)-1 > 31 (c.f. C99 6.5.7(3))
and UBSAN reports

  UBSAN: Undefined behaviour in fs/ext4/mballoc.c:2621:15
  shift exponent 4294967295 is too large for 32-bit type 'int'
  [...]
  Call Trace:
   [<ffffffff818c4d25>] dump_stack+0xbc/0x117
   [<ffffffff818c4c69>] ? _atomic_dec_and_lock+0x169/0x169
   [<ffffffff819411ab>] ubsan_epilogue+0xd/0x4e
   [<ffffffff81941cac>] __ubsan_handle_shift_out_of_bounds+0x1fb/0x254
   [<ffffffff81941ab1>] ? __ubsan_handle_load_invalid_value+0x158/0x158
   [<ffffffff814b6dc1>] ? kmem_cache_alloc+0x101/0x390
   [<ffffffff816fc13b>] ? ext4_mb_init+0x13b/0xfd0
   [<ffffffff814293c7>] ? create_cache+0x57/0x1f0
   [<ffffffff8142948a>] ? create_cache+0x11a/0x1f0
   [<ffffffff821c2168>] ? mutex_lock+0x38/0x60
   [<ffffffff821c23ab>] ? mutex_unlock+0x1b/0x50
   [<ffffffff814c26ab>] ? put_online_mems+0x5b/0xc0
   [<ffffffff81429677>] ? kmem_cache_create+0x117/0x2c0
   [<ffffffff816fcc49>] ext4_mb_init+0xc49/0xfd0
   [...]

Observe that the mentioned shift exponent, 4294967295, equals (unsigned)-1.

Unless compilers start to do some fancy transformations (which at least
GCC 6.0.0 doesn't currently do), the issue is of cosmetic nature only: the
such calculated value of offset is never used again.

Silence UBSAN by introducing another variable, offset_incr, holding the
next increment to apply to offset and adjust that one by right shifting it
by one position per loop iteration.

Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=114701
Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=112161

Signed-off-by: Nicolai Stange <nicstange@gmail.com>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

---
 fs/ext4/mballoc.c |    6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

--- a/fs/ext4/mballoc.c
+++ b/fs/ext4/mballoc.c
@@ -2578,7 +2578,7 @@ int ext4_mb_init(struct super_block *sb)
 {
 	struct ext4_sb_info *sbi = EXT4_SB(sb);
 	unsigned i, j;
-	unsigned offset;
+	unsigned offset, offset_incr;
 	unsigned max;
 	int ret;
 
@@ -2607,11 +2607,13 @@ int ext4_mb_init(struct super_block *sb)
 
 	i = 1;
 	offset = 0;
+	offset_incr = 1 << (sb->s_blocksize_bits - 1);
 	max = sb->s_blocksize << 2;
 	do {
 		sbi->s_mb_offsets[i] = offset;
 		sbi->s_mb_maxs[i] = max;
-		offset += 1 << (sb->s_blocksize_bits - i);
+		offset += offset_incr;
+		offset_incr = offset_incr >> 1;
 		max = max >> 1;
 		i++;
 	} while (i <= sb->s_blocksize_bits + 1);

  parent reply	other threads:[~2016-06-05 22:24 UTC|newest]

Thread overview: 100+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-06-05 21:40 [PATCH 4.4 00/99] 4.4.13-stable review Greg Kroah-Hartman
2016-06-05 21:40 ` [PATCH 4.4 01/99] MIPS64: R6: R2 emulation bugfix Greg Kroah-Hartman
2016-06-05 21:40 ` [PATCH 4.4 02/99] MIPS: math-emu: Fix jalr emulation when rd == $0 Greg Kroah-Hartman
2016-06-05 21:40 ` [PATCH 4.4 03/99] MIPS: MSA: Fix a link error on `_init_msa_upper with older GCC Greg Kroah-Hartman
2016-06-05 21:40 ` [PATCH 4.4 04/99] MIPS: Dont unwind to user mode with EVA Greg Kroah-Hartman
2016-06-05 21:40 ` [PATCH 4.4 05/99] MIPS: Avoid using unwind_stack() with usermode Greg Kroah-Hartman
2016-06-05 21:40 ` [PATCH 4.4 06/99] MIPS: Fix siginfo.h to use strict posix types Greg Kroah-Hartman
2016-06-05 21:40 ` [PATCH 4.4 07/99] MIPS: Fix uapi include in exported asm/siginfo.h Greg Kroah-Hartman
2016-06-05 21:40 ` [PATCH 4.4 08/99] MIPS: Fix watchpoint restoration Greg Kroah-Hartman
2016-06-05 21:40 ` [PATCH 4.4 09/99] MIPS: Handle highmem pages in __update_cache Greg Kroah-Hartman
2016-06-05 21:40 ` [PATCH 4.4 10/99] MIPS: Sync icache & dcache in set_pte_at Greg Kroah-Hartman
2016-06-05 21:40 ` [PATCH 4.4 11/99] MIPS: ath79: make bootconsole wait for both THRE and TEMT Greg Kroah-Hartman
2016-06-05 21:40 ` [PATCH 4.4 12/99] MIPS: Reserve nosave data for hibernation Greg Kroah-Hartman
2016-06-05 21:40 ` [PATCH 4.4 13/99] MIPS: Loongson-3: Reserve 32MB for RS780E integrated GPU Greg Kroah-Hartman
2016-06-05 21:40 ` [PATCH 4.4 14/99] MIPS: Use copy_s.fmt rather than copy_u.fmt Greg Kroah-Hartman
2016-06-05 21:40 ` [PATCH 4.4 15/99] MIPS: Fix MSA ld_*/st_* asm macros to use PTR_ADDU Greg Kroah-Hartman
2016-06-05 21:40 ` [PATCH 4.4 16/99] MIPS: Prevent "restoration" of MSA context in non-MSA kernels Greg Kroah-Hartman
2016-06-05 21:40 ` [PATCH 4.4 17/99] MIPS: Disable preemption during prctl(PR_SET_FP_MODE, ...) Greg Kroah-Hartman
2016-06-05 21:40 ` [PATCH 4.4 18/99] MIPS: ptrace: Fix FP context restoration FCSR regression Greg Kroah-Hartman
2016-06-05 21:40 ` [PATCH 4.4 19/99] MIPS: ptrace: Prevent writes to read-only FCSR bits Greg Kroah-Hartman
2016-06-05 21:40 ` [PATCH 4.4 20/99] MIPS: Fix sigreturn via VDSO on microMIPS kernel Greg Kroah-Hartman
2016-06-05 21:40 ` [PATCH 4.4 21/99] MIPS: Build microMIPS VDSO for microMIPS kernels Greg Kroah-Hartman
2016-06-05 21:40 ` [PATCH 4.4 22/99] MIPS: lib: Mark intrinsics notrace Greg Kroah-Hartman
2016-06-05 21:40 ` [PATCH 4.4 23/99] MIPS: VDSO: Build with `-fno-strict-aliasing Greg Kroah-Hartman
2016-06-05 21:40 ` [PATCH 4.4 24/99] affs: fix remount failure when there are no options changed Greg Kroah-Hartman
2016-06-05 21:40 ` [PATCH 4.4 25/99] ASoC: ak4642: Enable cache usage to fix crashes on resume Greg Kroah-Hartman
2016-06-05 21:40 ` [PATCH 4.4 26/99] Input: uinput - handle compat ioctl for UI_SET_PHYS Greg Kroah-Hartman
2016-06-05 21:41 ` [PATCH 4.4 27/99] ARM: mvebu: fix GPIO config on the Linksys boards Greg Kroah-Hartman
2016-06-05 21:41 ` [PATCH 4.4 28/99] ARM: dts: at91: fix typo in sama5d2 PIN_PD24 description Greg Kroah-Hartman
2016-06-05 21:41 ` [PATCH 4.4 29/99] ARM: dts: exynos: Add interrupt line to MAX8997 PMIC on exynos4210-trats Greg Kroah-Hartman
2016-06-05 21:41 ` [PATCH 4.4 30/99] ARM: dts: imx35: restore existing used clock enumeration Greg Kroah-Hartman
2016-06-05 21:41 ` [PATCH 4.4 31/99] ath9k: Add a module parameter to invert LED polarity Greg Kroah-Hartman
2016-06-05 21:41 ` [PATCH 4.4 32/99] ath9k: Fix LED polarity for some Mini PCI AR9220 MB92 cards Greg Kroah-Hartman
2016-06-05 21:41 ` [PATCH 4.4 33/99] ath10k: fix debugfs pktlog_filter write Greg Kroah-Hartman
2016-06-05 21:41 ` [PATCH 4.4 34/99] ath10k: fix firmware assert in monitor mode Greg Kroah-Hartman
2016-06-05 21:41 ` [PATCH 4.4 35/99] ath10k: fix rx_channel during hw reconfigure Greg Kroah-Hartman
2016-06-05 21:41 ` [PATCH 4.4 36/99] ath10k: fix kernel panic, move arvifs list head init before htt init Greg Kroah-Hartman
2016-06-05 21:41 ` [PATCH 4.4 37/99] ath5k: Change led pin configuration for compaq c700 laptop Greg Kroah-Hartman
2016-06-05 21:41 ` [PATCH 4.4 38/99] hwrng: exynos - Fix unbalanced PM runtime put on timeout error path Greg Kroah-Hartman
2016-06-05 21:41 ` [PATCH 4.4 39/99] rtlwifi: rtl8723be: Add antenna select module parameter Greg Kroah-Hartman
2016-06-05 21:41 ` [PATCH 4.4 40/99] rtlwifi: btcoexist: Implement antenna selection Greg Kroah-Hartman
2016-06-05 21:41 ` [PATCH 4.4 41/99] rtlwifi: Fix logic error in enter/exit power-save mode Greg Kroah-Hartman
2016-06-05 21:41 ` [PATCH 4.4 42/99] rtlwifi: pci: use dev_kfree_skb_irq instead of kfree_skb in rtl_pci_reset_trx_ring Greg Kroah-Hartman
2016-06-05 21:41 ` [PATCH 4.4 43/99] aacraid: Relinquish CPU during timeout wait Greg Kroah-Hartman
2016-06-05 21:41 ` [PATCH 4.4 44/99] aacraid: Fix for aac_command_thread hang Greg Kroah-Hartman
2016-06-05 21:41 ` [PATCH 4.4 45/99] aacraid: Fix for KDUMP driver hang Greg Kroah-Hartman
2016-06-05 21:41 ` [PATCH 4.4 46/99] regulator: Try to resolve regulators supplies on registration Greg Kroah-Hartman
2016-06-05 21:41 ` [PATCH 4.4 47/99] hwmon: (ads7828) Enable internal reference Greg Kroah-Hartman
2016-06-05 21:41 ` [PATCH 4.4 48/99] mfd: intel-lpss: Save register context on suspend Greg Kroah-Hartman
2016-06-05 21:41 ` [PATCH 4.4 50/99] PM / Runtime: Fix error path in pm_runtime_force_resume() Greg Kroah-Hartman
2016-06-05 21:41 ` [PATCH 4.4 51/99] cpuidle: Indicate when a device has been unregistered Greg Kroah-Hartman
2016-06-05 21:41 ` [PATCH 4.4 52/99] cpuidle: Fix cpuidle_state_is_coupled() argument in cpuidle_enter() Greg Kroah-Hartman
2016-06-05 21:41 ` [PATCH 4.4 53/99] clk: bcm2835: Fix PLL poweron Greg Kroah-Hartman
2016-06-05 21:41 ` [PATCH 4.4 54/99] clk: at91: fix check of clk_register() returned value Greg Kroah-Hartman
2016-06-05 21:41 ` [PATCH 4.4 55/99] clk: bcm2835: pll_off should only update CM_PLL_ANARST Greg Kroah-Hartman
2016-06-05 21:41 ` [PATCH 4.4 56/99] clk: bcm2835: divider value has to be 1 or more Greg Kroah-Hartman
2016-06-05 21:41 ` [PATCH 4.4 57/99] pinctrl: exynos5440: Use off-stack memory for pinctrl_gpio_range Greg Kroah-Hartman
2016-06-05 21:41 ` [PATCH 4.4 58/99] PCI: Disable all BAR sizing for devices with non-compliant BARs Greg Kroah-Hartman
2016-06-05 21:41 ` [PATCH 4.4 59/99] [media] media: v4l2-compat-ioctl32: fix missing reserved field copy in put_v4l2_create32 Greg Kroah-Hartman
2016-06-05 21:41 ` [PATCH 4.4 60/99] mm: use phys_addr_t for reserve_bootmem_region() arguments Greg Kroah-Hartman
2016-06-05 21:41 ` [PATCH 4.4 61/99] wait/ptrace: assume __WALL if the child is traced Greg Kroah-Hartman
2016-06-05 21:41 ` [PATCH 4.4 62/99] QE-UART: add "fsl,t1040-ucc-uart" to of_device_id Greg Kroah-Hartman
2016-06-05 21:41 ` [PATCH 4.4 63/99] [media] usbvision fix overflow of interfaces array Greg Kroah-Hartman
2016-06-05 21:53   ` Holger Hoffstätte
2016-06-08  0:26     ` Greg KH
2016-06-05 21:41 ` [PATCH 4.4 64/99] pipe: limit the per-user amount of pages allocated in pipes Greg Kroah-Hartman
2016-06-05 21:41 ` [PATCH 4.4 65/99] powerpc/book3s64: Fix branching to OOL handlers in relocatable kernel Greg Kroah-Hartman
2016-06-05 21:41 ` [PATCH 4.4 66/99] powerpc/eeh: Dont report error in eeh_pe_reset_and_recover() Greg Kroah-Hartman
2016-06-05 21:41 ` [PATCH 4.4 67/99] Revert "powerpc/eeh: Fix crash in eeh_add_device_early() on Cell" Greg Kroah-Hartman
2016-06-05 21:41 ` [PATCH 4.4 68/99] powerpc/eeh: Restore initial state in eeh_pe_reset_and_recover() Greg Kroah-Hartman
2016-06-05 21:41 ` [PATCH 4.4 69/99] xen/events: Dont move disabled irqs Greg Kroah-Hartman
2016-06-05 21:41 ` [PATCH 4.4 70/99] xen: use same main loop for counting and remapping pages Greg Kroah-Hartman
2016-06-05 21:41 ` [PATCH 4.4 72/99] drm/gma500: Fix possible out of bounds read Greg Kroah-Hartman
2016-06-05 21:41 ` [PATCH 4.4 73/99] drm/vmwgfx: Enable SVGA_3D_CMD_DX_SET_PREDICATION Greg Kroah-Hartman
2016-06-05 21:41 ` [PATCH 4.4 74/99] drm/vmwgfx: use vmw_cmd_dx_cid_check for query commands Greg Kroah-Hartman
2016-06-05 21:41 ` [PATCH 4.4 75/99] drm/vmwgfx: Fix order of operation Greg Kroah-Hartman
2016-06-05 21:41 ` [PATCH 4.4 76/99] drm/amdgpu: use drm_mode_vrefresh() rather than mode->vrefresh Greg Kroah-Hartman
2016-06-05 21:41 ` [PATCH 4.4 77/99] drm/amdgpu: Fix hdmi deep color support Greg Kroah-Hartman
2016-06-05 21:41 ` [PATCH 4.4 78/99] drm/i915/fbdev: Fix num_connector references in intel_fb_initial_config() Greg Kroah-Hartman
2016-06-05 21:41 ` [PATCH 4.4 79/99] drm/fb_helper: Fix references to dev->mode_config.num_connector Greg Kroah-Hartman
2016-06-05 21:41 ` [PATCH 4.4 80/99] drm/atomic: Verify connector->funcs != NULL when clearing states Greg Kroah-Hartman
2016-06-05 21:41 ` [PATCH 4.4 83/99] ext4: fix hang when processing corrupted orphaned inode list Greg Kroah-Hartman
2016-06-05 21:41 ` [PATCH 4.4 84/99] ext4: clean up error handling when orphan list is corrupted Greg Kroah-Hartman
2016-06-05 21:41 ` [PATCH 4.4 85/99] ext4: fix oops on corrupted filesystem Greg Kroah-Hartman
2016-06-05 21:41 ` [PATCH 4.4 86/99] ext4: address UBSAN warning in mb_find_order_for_block() Greg Kroah-Hartman
2016-06-05 21:42 ` Greg Kroah-Hartman [this message]
2016-06-05 21:42 ` [PATCH 4.4 88/99] PM / sleep: Handle failures in device_suspend_late() consistently Greg Kroah-Hartman
2016-06-05 21:42 ` [PATCH 4.4 90/99] scripts/package/Makefile: rpmbuild add support of RPMOPTS Greg Kroah-Hartman
2016-06-05 21:42 ` [PATCH 4.4 91/99] gcov: disable tree-loop-im to reduce stack usage Greg Kroah-Hartman
2016-06-05 21:42 ` [PATCH 4.4 92/99] xfs: disallow rw remount on fs with unknown ro-compat features Greg Kroah-Hartman
2016-06-05 21:42 ` [PATCH 4.4 93/99] xfs: Dont wrap growfs AGFL indexes Greg Kroah-Hartman
2016-06-05 21:42 ` [PATCH 4.4 94/99] xfs: xfs_iflush_cluster fails to abort on error Greg Kroah-Hartman
2016-06-05 21:42 ` [PATCH 4.4 95/99] xfs: fix inode validity check in xfs_iflush_cluster Greg Kroah-Hartman
2016-06-05 21:42 ` [PATCH 4.4 96/99] xfs: skip stale inodes " Greg Kroah-Hartman
2016-06-05 21:42 ` [PATCH 4.4 98/99] xfs: handle dquot buffer readahead in log recovery correctly Greg Kroah-Hartman
2016-06-05 21:42 ` [PATCH 4.4 99/99] gpio: davinci: fix missed parent conversion Greg Kroah-Hartman
2016-06-06 17:26 ` [PATCH 4.4 00/99] 4.4.13-stable review Shuah Khan
2016-06-07  4:07 ` Guenter Roeck
2016-06-08  1:02   ` Greg Kroah-Hartman
2016-06-08  1:07   ` Greg Kroah-Hartman

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=20160605213911.491082493@linuxfoundation.org \
    --to=gregkh@linuxfoundation.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=nicstange@gmail.com \
    --cc=stable@vger.kernel.org \
    --cc=tytso@mit.edu \
    /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).