Quantal update to 3.5.7.9 stable release

SRU Justification

    Impact:
       The upstream process for stable tree updates is quite similar
       in scope to the Ubuntu SRU process, e.g., each patch has to
       demonstrably fix a bug, and each patch is vetted by upstream
       by originating either directly from Linus' tree or in a minimally
       backported form of that patch. The 3.5.7.8 and 3.5.7.9 upstream stable
       patch sets are now available.They should be included in the Ubuntu
       kernel as well.

       git://kernel.ubuntu.com/ubuntu/linux.git

    TEST CASE: TBD

       The following patches are in the 3.2.41 and 3.2.42 stable releases:

 * Linux 3.5.7.9
 * vfs,proc: guarantee unique inodes in /proc
 * kernel/signal.c: use __ARCH_HAS_SA_RESTORER instead of SA_RESTORER
 * signal: Define __ARCH_HAS_SA_RESTORER so we know whether to clear sa_restorer
 * bnx2x: fix occasional statistics off-by-4GB error
 * inet: limit length of fragment queue hash table bucket lists
 * tcp: fix skb_availroom()
 * rtnetlink: Mask the rta_type when range checking
 * ipv4: fix definition of FIB_TABLE_HASHSZ
 * sctp: Use correct sideffect command in duplicate cookie handling
 * bonding: don't call update_speed_duplex() under spinlocks
 * net/ipv4: Ensure that location of timestamp option is stored
 * netconsole: don't call __netpoll_cleanup() while atomic
 * sctp: don't break the loop while meeting the active_path so as to find the matched transport
 * sfc: Disable soft interrupt handling during efx_device_detach_sync()
 * sfc: Only use TX push if a single descriptor is to be written
 * sfc: Detach net device when stopping queues for reconfiguration
 * sfc: Correct efx_rx_buffer::page_offset when EFX_PAGE_IP_ALIGN != 0
 * sfc: Fix efx_rx_buf_offset() in the presence of swiotlb
 * sfc: Properly sync RX DMA buffer when it is not the last in the page
 * sfc: Fix timekeeping in efx_mcdi_poll()
 * sfc: Correctly initialise reset_method in siena_test_chip()
 * sfc: lock TX queues when calling netif_device_detach()
 * sfc: Work-around flush timeout when flushes have completed
 * sfc: Avoid generating over-length MC_CMD_FLUSH_RX_QUEUES request
 * sfc: Fix MCDI structure field lookup
 * sfc: Add parentheses around use of bitfield macro arguments
 * sfc: Convert firmware subtypes to native byte order in efx_mcdi_get_board_cfg()
 * sfc: Really disable flow control while flushing
 * sfc: Disable VF queues during register self-test
 * sunsu: Fix panic in case of nonexistent port at "console=ttySY" cmdline option
 * tg3: 5715 does not link up when autoneg off
 * KMS: fix EDID detailed timing frame rate
 * KMS: fix EDID detailed timing vsync parsing
 * drivers/rtc/rtc-at91rm9200.c: use a variable for storing IMR
 * mm/hugetlb: fix total hugetlbfs pages count when using memory overcommit accouting
 * i2c: tegra: check the clk_prepare_enable() return value
 * USB: serial: fix interface refcounting
 * USB: io_ti: fix get_icount for two port adapters
 * USB: garmin_gps: fix memory leak on disconnect
 * USB: cdc-acm: fix device unregistration
 * cifs: ignore everything in SPNEGO blob after mechTypes
 * drm/mgag200: Bug fix: Modified pll algorithm for EH project
 * dm verity: avoid deadlock
 * dm thin: fix discard corruption
 * usb: gadget: ffs: fix enable multiple instances
 * usb: gadget: udc-core: fix a regression during gadget driver unbinding
 * ALSA: hda - Fix typo in checking IEC958 emphasis bit
 * ext4: fix data=journal fast mount/umount hang
 * ALSA: snd-usb: mixer: ignore -EINVAL in snd_usb_mixer_controls()
 * ALSA: snd-usb: mixer: propagate errors up the call chain
 * target/file: Bump FD_MAX_SECTORS to 2048 to handle 1M sized I/Os
 * target/iscsi: Fix mutual CHAP auth on big-endian arches
 * x86-64: Fix the failure case in copy_user_handle_tail()
 * USB: xhci - fix bit definitions for IMAN register
 * ALSA: hda/cirrus - Fix the digital beep registration
 * drm/radeon: add Richland pci ids
 * drm/radeon: add support for Richland APUs
 * drm/radeon/benchmark: make sure bo blit copy exists before using it
 * drm/radeon: fix backend map setup on 1 RB trinity boards
 * usb-storage: add unusual_devs entry for Samsung YP-Z3 mp3 player
 * USB: xhci: correctly enable interrupts
 * tracing: Keep overwrite in sync between regular and snapshot buffers
 * tracing: Fix free of probe entry by calling call_rcu_sched()
 * drm/i915: bounds check execbuffer relocation count
 * drm/i915: restrict kernel address leak in debugfs
 * cifs: delay super block destruction until all cifsFileInfo objects are gone
 * tracing: Fix race in snapshot swapping
 * ext4: use atomic64_t for the per-flexbg free_clusters count
 * jbd2: fix use after free in jbd2_journal_dirty_metadata()
 * ext4: fix the wrong number of the allocated blocks in ext4_split_extent()
 * key: Fix resource leak
 * isofs: avoid info leak on export
 * udf: avoid info leak on export
 * vhost/net: fix heads usage of ubuf_info
 * rtlwifi: rtl8192cu: Fix problem that prevents reassociation
 * rtlwifi: rtl8192cu: Fix schedule while atomic bug splat
 * mwifiex: fix potential out-of-boundary access to ibss rate table
 * rt2x00: error in configurations with mesh support disabled
 * HID: clean up quirk for Sony RF receivers
 * HID: add support for Sony RF receiver with USB product id 0x0374
 * ahci: Add Device IDs for Intel Wellsburg PCH
 * ahci: AHCI-mode SATA patch for Intel Avoton DeviceIDs
 * ahci: Add Device IDs for Intel Lynx Point-LP PCH
 * 6lowpan: Fix endianness issue in is_addr_link_local().
 * dcbnl: fix various netlink info leaks
 * rtnl: fix info leak on RTM_GETLINK request for VF devices
 * ipv6: stop multicast forwarding to process interface scoped addresses
 * bridging: fix rx_handlers return code
 * netlabel: fix build problems when CONFIG_IPV6=n
 * netlabel: correctly list all the static label mappings
 * macvlan: Set IFF_UNICAST_FLT flag to prevent unnecessary promisc mode.
 * tun: add a missing nf_reset() in tun_net_xmit()
 * tcp: fix double-counted receiver RTT when leaving receiver fast path
 * net: ipv6: Don't purge default router if accept_ra=2
 * rds: limit the size allocated by rds_message_alloc()
 * l2tp: Restore socket refcount when sendmsg succeeds
 * perf,x86: fix wrmsr_on_cpu() warning on suspend/resume
 * perf,x86: fix link failure for non-Intel configs
 * perf,x86: fix kernel crash with PEBS/BTS after suspend/resume
 * selinux: use GFP_ATOMIC under spin_lock
 * atmel_lcdfb: fix 16-bpp modes on older SOCs
 * powerpc: Fix cputable entry for 970MP rev 1.0
 * powerpc: Fix STAB initialization
 * s390/mm: fix flush_tlb_kernel_range()
 * s390: critical section cleanup vs. machine checks
 * proc connector: reject unprivileged listener bumps
 * loopdev: remove an user triggerable oops
 * loopdev: fix a deadlock
 * block: use i_size_write() in bd_set_size()
 * virtio: rng: disallow multiple device registrations, fixes crashes
 * TTY: do not reset master's packet mode
 * hwmon: (sht15) Fix memory leak if regulator_enable() fails
 * btrfs: use rcu_barrier() to wait for bdev puts at unmount
 * hwmon: (pmbus/ltc2978) Fix temperature reporting
 * hwmon: (lineage-pem) Add missing terminating entry for pem_[input|fan]_attributes
 * USB: cdc-wdm: fix buffer overflow
 * w1: fix oops when w1_search is called from netlink connector
 * usb: serial: Add Rigblaster Advantage to device table
 * tty: serial: fix typo "ARCH_S5P6450"
 * serial: 8250_pci: add support for another kind of NetMos Technology PCI 9835 Multi-I/O Controller
 * Fix 4 port and add support for 8 port 'Unknown' PCI serial port cards
 * tty/serial: Add support for Altera serial port
 * ext3: Fix format string issues
 * qcaux: add Franklin U600
 * staging: vt6656: Fix oops on resume from suspend.
 * usb: cp210x new Vendor/Device IDs
 * ALSA: seq: Fix missing error handling in snd_seq_timer_open()
 * USB: added support for Cinterion's products AH6 and PLS8
 * USB: storage: fix Huawei mode switching regression
 * USB: option: add Huawei E5331
 * Linux 3.5.7.8
 * USB: Fix connected device switch to Inactive state.
 * USB: Rip out recursive call on warm port reset.
 * USB: Prepare for refactoring by adding extra udev checks.
 * USB: Don't use EHCI port sempahore for USB 3.0 hubs.
 * USB: Handle warm reset failure on empty port.
 * drm/i915: Increase the RC6p threshold.
 * Fix: compat_rw_copy_check_uvector() misuse in aio, readv, writev, and security keys
 * vfs: fix pipe counter breakage
 * xen/pciback: Don't disable a PCI device that is already disabled.
 * Btrfs: fix panic when recovering tree log
 * Fix memory leak in cpufreq stats.
 * e1000e: fix pci-device enable-counter balance
 * ftrace: Update the kconfig for DYNAMIC_FTRACE
 * HID: logitech-dj: do not directly call hid_output_raw_report() during probe
 * Btrfs: account for orphan inodes properly during cleanup
 * dmi_scan: fix missing check for _DMI_ signature in smbios_present()
 * drm/radeon: add primary dac adj quirk for R200 board
 * ALSA: ice1712: Initialize card->private_data properly
 * ALSA: vmaster: Fix slave change notification
 * random: fix locking dependency with the tasklist_lock
 * hwmon: (sht15) Check return value of regulator_enable()
 * hwmon: (pmbus/ltc2978) Use detected chip ID to select supported functionality
 * hwmon: (pmbus/ltc2978) Fix peak attribute handling
 * userns: Stop oopsing in key_change_session_keyring
 * ARM: 7663/1: perf: fix ARMv7 EVTYPE_MASK to include NSH bit
 * ARM: 7657/1: head: fix swapper and idmap population with LPAE and big-endian
 * drm/i915: Don't clobber crtc->fb when queue_flip fails
 * hw_random: make buffer usable in scatterlist.
 * libertas: fix crash for SD8688
 * mwifiex: correct sleep delay counter
 * ath9k_hw: improve reset reliability after errors
 * ath9k_htc: fix signal strength handling issues
 * ath9k: fix RSSI dummy marker value
 * md: raid0: fix error return from create_stripe_zones.
 * md: fix two bugs when attempting to resize RAID0 array.
 * md: protect against crash upon fsync on ro array
 * zram: Fix deadlock bug in partial read/write
 * workqueue: consider work function when searching for busy work items
 * unbreak automounter support on 64-bit kernel with 32-bit userspace (v2)
 * pstore: Avoid deadlock in panic and emergency-restart path
 * usb hid quirks for Masterkit MA901 usb radio
 * fuse: don't WARN when nlink is zero
 * fs: cachefiles: add support for large files in filesystem caching
 * ata_piix: Add Device IDs for Intel Lynx Point-LP PCH
 * ata_piix: Add Device IDs for Intel Wellsburg PCH
 * ata_piix: IDE-mode SATA patch for Intel Avoton DeviceIDs
 * asus-laptop: Do not call HWRS on init
 * nfsd: add get_uint for u32's
 * ata_piix: reenable MS Virtual PC guests
 * ext4: convert number of blocks to clusters properly
 * dm snapshot: add missing module aliases
 * xenbus: fix compile failure on ARM with Xen enabled
 * xen/pci: We don't do multiple MSI's.
 * xen/pat: Disable PAT using pat_enabled value.
 * target/pscsi: Fix page increment
 * nbd: fsync and kill block device on shutdown
 * sysctl: fix null checking in bin_dn_node_address()
 * idr: fix top layer handling
 * firewire: add minor number range check to fw_device_init()
 * block: fix synchronization and limit check in blk_alloc_devt()
 * idr: fix a subtle bug in idr_get_next()
 * block: fix ext_devt_idr handling
 * ocfs2: ac->ac_allow_chain_relink=0 won't disable group relink
 * ocfs2: fix ocfs2_init_security_and_acl() to initialize acl correctly
 * iscsi-target: Fix immediate queue starvation regression with DATAIN
 * cifs: ensure that cifs_get_root() only traverses directories
 * x86: Make sure we can boot in the case the BDA contains pure garbage
 * sony-laptop: fully enable SNY controlled modems
 * doc, kernel-parameters: Document 'console=hvc<n>'
 * doc, xen: Mention 'earlyprintk=xen' in the documentation.
 * ARM: fix scheduling while atomic warning in alignment handling code
 * ARM: VFP: fix emulation of second VFP instruction
 * ALSA: bt87x: Make load_all parameter working again
 * dc395x: uninitialized variable in device_alloc()
 * storvsc: Initialize the sglist
 * SUNRPC: Don't start the retransmission timer when out of socket space
 * NFS: Don't allow NFS silly-renamed files to be deleted, no signal
 * btrfs: Init io_lock after cloning btrfs device struct
 * xen-blkback: do not leak mode property
 * target: Add missing mapped_lun bounds checking during make_mappedlun setup
 * svcrpc: make svc_age_temp_xprts enqueue under sv_lock
 * iommu/amd: Initialize device table after dma_ops
 * nfsd: Fix memleak
 * omap_vout: find_vma() needs ->mmap_sem held

    The following patches had already been applied:

 * exec: use -ELOOP for max recursion depth
 * vhost: fix length for cross region descriptor
 * keys: fix race with concurrent
 * crypto: user - fix info leaks in report API
 * signal: always clear sa_restorer on execve
 * mm/hotplug: correctly add new zone to all other
 * tools: hv: Netlink source address validation allows