kernel-packages team mailing list archive

[Launchpad Bug Tracker <1226726@xxxxxxxxxxxxxxxxxx>]

This bug was fixed in the package linux - 3.8.0-33.48

---------------
linux (3.8.0-33.48) raring; urgency=low

  [ Brad Figg ]

  * Release Tracking Bug
    - LP: #1242849

  [ Maximiliano Curia ]

  * SAUCE: (no-up) Only let characters through when there are active
    readers.
    - LP: #1208740

  [ Upstream Kernel Changes ]

  * cciss: fix info leak in cciss_ioctl32_passthru()
    - LP: #1188355
    - CVE-2013-2147
  * cpqarray: fix info leak in ida_locked_ioctl()
    - LP: #1188355
    - CVE-2013-2147
  * mount: consolidate permission checks
    - LP: #1226726
  * get rid of full-hash scan on detaching vfsmounts
    - LP: #1226726
  * Smack: Fix the bug smackcipso can't set CIPSO correctly
    - LP: #1236743
  * ipvs: add backup_only flag to avoid loops
    - LP: #1238494
  * tuntap: correctly handle error in tun_set_iff()
    - LP: #1229975
    - CVE-2013-4343
  * htb: fix sign extension bug
    - LP: #1240580
  * net: avoid to hang up on sending due to sysctl configuration overflow.
    - LP: #1240580
  * net: check net.core.somaxconn sysctl values
    - LP: #1240580
  * macvlan: validate flags
    - LP: #1240580
  * neighbour: populate neigh_parms on alloc before calling ndo_neigh_setup
    - LP: #1240580
  * bonding: modify only neigh_parms owned by us
    - LP: #1240580
  * fib_trie: remove potential out of bound access
    - LP: #1240580
  * bridge: don't try to update timers in case of broken MLD queries
    - LP: #1240580
  * tcp: cubic: fix overflow error in bictcp_update()
    - LP: #1240580
  * tcp: cubic: fix bug in bictcp_acked()
    - LP: #1240580
  * ipv6: don't stop backtracking in fib6_lookup_1 if subtree does not
    match
    - LP: #1240580
  * 8139cp: Fix skb leak in rx_status_loop failure path.
    - LP: #1240580
  * tun: signedness bug in tun_get_user()
    - LP: #1240580
  * ipv6: remove max_addresses check from ipv6_create_tempaddr
    - LP: #1240580
  * ipv6: Store Router Alert option in IP6CB directly.
    - LP: #1240580
  * ipv6: drop packets with multiple fragmentation headers
    - LP: #1240580
  * tcp: set timestamps for restored skb-s
    - LP: #1240580
  * net: usb: Add HP hs2434 device to ZLP exception table
    - LP: #1240580
  * tcp: initialize rcv_tstamp for restored sockets
    - LP: #1240580
  * ipv4: sendto/hdrincl: don't use destination address found in header
    - LP: #1240580
  * tcp: tcp_make_synack() should use sock_wmalloc
    - LP: #1240580
  * tipc: set sk_err correctly when connection fails
    - LP: #1240580
  * net: bridge: convert MLDv2 Query MRC into msecs_to_jiffies for
    max_delay
    - LP: #1240580
  * ICMPv6: treat dest unreachable codes 5 and 6 as EACCES, not EPROTO
    - LP: #1240580
  * tg3: Don't turn off led on 5719 serdes port 0
    - LP: #1240580
  * vhost_net: poll vhost queue after marking DMA is done
    - LP: #1240580
  * net: ipv6: tcp: fix potential use after free in tcp_v6_do_rcv
    - LP: #1240580
  * drm/radeon/si: Add support for CP DMA to CS checker for compute v2
    - LP: #1240580
  * sfc: Fix efx_rx_buf_offset() for recycled pages
    - LP: #1240580
  * cfq: explicitly use 64bit divide operation for 64bit arguments
    - LP: #1240580
  * drm/radeon/atom: workaround vbios bug in transmitter table on rs880
    (v2)
    - LP: #1240580
  * drm/ast: fix the ast open key function
    - LP: #1240580
  * sched/fair: Fix small race where child->se.parent,cfs_rq might point to
    invalid ones
    - LP: #1240580
  * tg3: Expand led off fix to include 5720
    - LP: #1240580
  * HID: provide a helper for validating hid reports
    - LP: #1240580
  * HID: zeroplus: validate output report details
    - LP: #1240580
    - CVE-2013-2889
  * HID: LG: validate HID output report details
    - LP: #1240580
    - CVE-2013-2893
  * HID: lenovo-tpkbd: validate output report details
    - LP: #1240580
    - CVE-2013-2894
  * HID: validate feature and input report details
    - LP: #1240580
    - CVE-2013-2897
  * HID: logitech-dj: validate output report details
    - LP: #1240580
    - CVE-2013-2895
  * HID: multitouch: validate indexes details
    - LP: #1240580
    - CVE-2013-2897
  * HID: lenovo-tpkbd: fix leak if tpkbd_probe_tp fails
    - LP: #1240580
  * drm/radeon: fix panel scaling with eDP and LVDS bridges
    - LP: #1240580
  * cifs: fix filp leak in cifs_atomic_open()
    - LP: #1240580
  * net: usb: cdc_ether: Use wwan interface for Telit modules
    - LP: #1240580
  * usb: gadget: fix a bug and a WARN_ON in dummy-hcd
    - LP: #1240580
  * drm/i915: do not update cursor in crtc mode set
    - LP: #1240580
  * drm/i915: Don't enable the cursor on a disable pipe
    - LP: #1240580
  * drm/ttm: fix the tt_populated check in ttm_tt_destroy()
    - LP: #1240580
  * PCI / ACPI / PM: Clear pme_poll for devices in D3cold on wakeup
    - LP: #1240580
  * serial: pch_uart: fix tty-kref leak in dma-rx path
    - LP: #1240580
  * x86, efi: Don't map Boot Services on i386
    - LP: #1240580
  * ALSA: compress: Fix compress device unregister.
    - LP: #1240580
  * dm snapshot: workaround for a false positive lockdep warning
    - LP: #1240580
  * dm-snapshot: fix performance degradation due to small hash size
    - LP: #1240580
  * drm/radeon: Make r100_cp_ring_info() and radeon_ring_gfx() safe (v2)
    - LP: #1240580
  * ARM: 7837/3: fix Thumb-2 bug in AES assembler code
    - LP: #1240580
  * x86/reboot: Add quirk to make Dell C6100 use reboot=pci automatically
    - LP: #1240580
  * drm/radeon: disable tests/benchmarks if accel is disabled
    - LP: #1240580
  * xhci: Fix oops happening after address device timeout
    - LP: #1240580
  * xhci: Ensure a command structure points to the correct trb on the
    command ring
    - LP: #1240580
  * drm/i915/dp: increase i2c-over-aux retry interval on AUX DEFER
    - LP: #1240580
  * staging: vt6656: [BUG] main_usb.c oops on device_close move flag
    earlier.
    - LP: #1240580
  * staging: vt6656: [BUG] iwctl_siwencodeext return if device not open
    - LP: #1240580
  * USB: UHCI: accept very late isochronous URBs
    - LP: #1240580
  * USB: OHCI: accept very late isochronous URBs
    - LP: #1240580
  * USB: fix PM config symbol in uhci-hcd, ehci-hcd, and xhci-hcd
    - LP: #1240580
  * usb/core/devio.c: Don't reject control message to endpoint with wrong
    direction bit
    - LP: #1240580
  * hwmon: (applesmc) Check key count before proceeding
    - LP: #1240580
  * fsl/usb: Resolve PHY_CLK_VLD instability issue for ULPI phy
    - LP: #1240580
  * driver core : Fix use after free of dev->parent in device_shutdown
    - LP: #1240580
  * USB: Fix breakage in ffs_fs_mount()
    - LP: #1240580
  * usb: dwc3: pci: add support for BayTrail
    - LP: #1240580
  * usb: dwc3: add support for Merrifield
    - LP: #1240580
  * ASoC: max98095: a couple array underflows
    - LP: #1240580
  * ASoC: ab8500-codec: info leak in anc_status_control_put()
    - LP: #1240580
  * ASoC: 88pm860x: array overflow in snd_soc_put_volsw_2r_st()
    - LP: #1240580
  * Bluetooth: Add a new PID/VID 0cf3/e005 for AR3012.
    - LP: #1240580
  * Bluetooth: Fix security level for peripheral role
    - LP: #1240580
  * Bluetooth: Fix encryption key size for peripheral role
    - LP: #1240580
  * Bluetooth: Add support for BCM20702A0 [0b05, 17cb]
    - LP: #1240580
  * Bluetooth: Introduce a new HCI_RFKILLED flag
    - LP: #1240580
  * rtlwifi: Align private space in rtl_priv struct
    - LP: #1240580
  * p54usb: add USB ID for Corega WLUSB2GTST USB adapter
    - LP: #1240580
  * mwifiex: fix hang issue for USB chipsets
    - LP: #1240580
  * mwifiex: fix NULL pointer dereference in usb suspend handler
    - LP: #1240580
  * fs/binfmt_elf.c: prevent a coredump with a large vm_map_count from
    Oopsing
    - LP: #1240580
  * nilfs2: fix issue with race condition of competition between segments
    for dirty blocks
    - LP: #1240580
  * mm: avoid reinserting isolated balloon pages into LRU lists
    - LP: #1240580
  * USB: serial: option: Ignore card reader interface on Huawei E1750
    - LP: #1240580
  * gpio/omap: maintain GPIO and IRQ usage separately
    - LP: #1240580
  * gpio/omap: auto-setup a GPIO when used as an IRQ
    - LP: #1240580
  * ib_srpt: Destroy cm_id before destroying QP.
    - LP: #1240580
  * powerpc: Fix parameter clobber in csum_partial_copy_generic()
    - LP: #1240580
  * powerpc: Restore registers on error exit from
    csum_partial_copy_generic()
    - LP: #1240580
  * powerpc/sysfs: Disable writing to PURR in guest mode
    - LP: #1240580
  * powerpc/iommu: Use GFP_KERNEL instead of GFP_ATOMIC in
    iommu_init_table()
    - LP: #1240580
  * powerpc/vio: Fix modalias_show return values
    - LP: #1240580
  * ib_srpt: always set response for task management
    - LP: #1240580
  * xen/hvc: allow xenboot console to be used again
    - LP: #1240580
  * net: Update the sysctl permissions handler to test effective uid/gid
    - LP: #1240580
  * Linux 3.8.13.11
    - LP: #1240580
 -- Brad Figg <brad.figg@xxxxxxxxxxxxx>   Mon, 21 Oct 2013 12:04:49 -0700

-- 
You received this bug notification because you are a member of Kernel
Packages, which is subscribed to linux in Ubuntu.
https://bugs.launchpad.net/bugs/1226726

Title:
  dentry_reset_mounted walks entire mount list holding vfsmount write
  lock

Status in “linux” package in Ubuntu:
  Fix Released
Status in “linux” source package in Precise:
  Won't Fix
Status in “linux” source package in Quantal:
  Fix Released
Status in “linux” source package in Raring:
  Fix Released
Status in “linux” source package in Saucy:
  Fix Released

Bug description:
  SRU Justification:

  Impact: When creating thousands of network namespaces the delay in
  executing commands increases exponentially in kernels before 84d17192.

  Fix: In 84d17192 in the upstream kernel, locking code in
  fs/namespace.c is greatly improved resulting in much better
  performance when the number namespaces increase.

  Testcase: Below, test_ns.sh can be run and a graph can be compared
  between the existing version and the patched version.

  Additional Information: Because this is a change in the vfs layer, I
  ran the xfstests and compared before and after results of this patch.
  The patch did not create any additional failures in the generic
  xfstests.

  The quantal and raring solutions differ but are both based on the 84d17192
  patch. The quantal solution does a backport of this patch instead of clean
  cherry-picks because of the amount of deps required to just use cherry-picks.
  The raring solution was able to be done with two clean cherry-picks and that's
  why that solution was chosen.

  --

  Whenever one enters a network namespace via "ip netns exec foobar
  somecommand" there is a mount done of the appropriate device on /sys
  since "somecommand" needs to see namespace specific versions of /sys
  directories. When the ip process exits these mounts need to be torn
  down, and that requires a global write lock for vfsmount_lock (this is
  a single writer multiple reader lock). This has serious performance
  implications when the number of name spaces increase.

  The commit 84d17192 addresses this issue, and it is clear by running
  the attached testcase that it fixes performance issues when dealing
  with large numbers of namespaces. I've included a graph with the
  differences in performance between this fix and its parent commit to
  show the the improve in performance. The x-axis represents the number
  of namespaces and the y-axis is execution time in ms. After applying
  the patch the performance delays are not exponentially increasing.

  This affects 3.2/3.5/3.8 series kernels, as it was fixed in 3.10.

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1226726/+subscriptions