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 3.2.40 upstream stable patch set is now available. It should be included in the Ubuntu kernel as well. git://git.kernel.org/ TEST CASE: TBD The following patches are in the 3.2.41 and 3.2.42 stable releases: * Linux 3.2.42 * 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 * efivars: pstore: Do not check size when erasing variable * KMS: fix EDID detailed timing frame rate * KMS: fix EDID detailed timing vsync parsing * mm/hugetlb: fix total hugetlbfs pages count when using memory overcommit accouting * vfs,proc: guarantee unique inodes in /proc * 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 * cifs: ignore everything in SPNEGO blob after mechTypes * efivars: Handle duplicate names from get_next_variable() * efivars: explicitly calculate length of VariableName * efi_pstore: Introducing workqueue updating sysfs * UBUNTU: SAUCE: Update configs for new efivars option * efivars: Fix check for CONFIG_EFI_VARS_PSTORE_DEFAULT_DISABLE * efivars: Add module parameter to disable use as a pstore backend * efivars: Allow disabling use as a pstore backend * dm thin: fix discard corruption * 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 * 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 * c drm/radeon/benchmark: make sure bo blit copy exists before using it * usb-storage: add unusual_devs entry for Samsung YP-Z3 mp3 player * USB: xhci: correctly enable interrupts * tracing: Prevent buffer overwrite disabled for latency tracers * tracing: Keep overwrite in sync between regular and snapshot buffers * tracing: Protect tracer flags with trace_types_lock * 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 * ext4: convert number of blocks to clusters properly * jbd2: fix use after free in jbd2_journal_dirty_metadata() * ext4: fix the wrong number of the allocated blocks in ext4_split_extent() * sfc: Only use TX push if a single descriptor is to be written * sfc: Disable soft interrupt handling during efx_device_detach_sync() * sfc: Detach net device when stopping queues for reconfiguration * 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: lock TX queues when calling netif_device_detach() * sfc: Fix two causes of flush failure * sfc: Convert firmware subtypes to native byte order in efx_mcdi_get_board_cfg() * sfc: Do not attempt to flush queues if DMA is disabled * inet: limit length of fragment queue hash table bucket lists * rtnetlink: Mask the rta_type when range checking * tcp: fix skb_availroom() * ipv4: fix definition of FIB_TABLE_HASHSZ * sctp: don't break the loop while meeting the active_path so as to find the matched transport * sctp: Use correct sideffect command in duplicate cookie handling * bonding: don't call update_speed_duplex() under spinlocks * netconsole: don't call __netpoll_cleanup() while atomic * net/ipv4: Ensure that location of timestamp option is stored * sunsu: Fix panic in case of nonexistent port at "console=ttySY" cmdline option * isofs: avoid info leak on export * udf: avoid info leak on export * selinux: use GFP_ATOMIC under spin_lock * vhost/net: fix heads usage of ubuf_info * rtlwifi: rtl8192cu: Fix problem that prevents reassociation * powerpc: Fix cputable entry for 970MP rev 1.0 * rtlwifi: rtl8192cu: Fix schedule while atomic bug splat * mwifiex: fix potential out-of-boundary access to ibss rate table * s390/mm: fix flush_tlb_kernel_range() * i915: initialize CADL in opregion * perf: Revert duplicated commit * drm/i915: Increase the RC6p threshold. * 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: 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 link failure for non-Intel configs * perf,x86: fix wrmsr_on_cpu() warning on suspend/resume * perf,x86: fix kernel crash with PEBS/BTS after suspend/resume * TTY: do not reset master's packet mode * Linux 3.2.41 * btrfs: use rcu_barrier() to wait for bdev puts at unmount * loopdev: remove an user triggerable oops * loopdev: fix a deadlock * block: use i_size_write() in bd_set_size() * batman-adv: Only write requested number of byte to user buffer * batman-adv: bat_socket_read missing checks * 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. * Fix: compat_rw_copy_check_uvector() misuse in aio, readv, writev, and security keys * 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() * virtio: rng: disallow multiple device registrations, fixes crashes * e1000e: fix pci-device enable-counter balance * USB: added support for Cinterion's products AH6 and PLS8 * USB: storage: fix Huawei mode switching regression * USB: option: add Huawei E5331 * xen/pciback: Don't disable a PCI device that is already disabled. * vfs: fix pipe counter breakage * Fix memory leak in cpufreq stats. * decnet: Fix disappearing sysctl entries * ftrace: Update the kconfig for DYNAMIC_FTRACE * efi: be more paranoid about available space when creating variables * efivars: Disable external interrupt while holding efivars->lock * efi_pstore: Check remaining space with QueryVariableInfo() before writing data * tty: Correct tty buffer flush. * iommu/amd: Initialize device table after dma_ops * 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 * HID: clean up quirk for Sony RF receivers * HID: add support for Sony RF receiver with USB product id 0x0374 * iwlwifi: always copy first 16 bytes of commands * dmi_scan: fix missing check for _DMI_ signature in smbios_present() * drm/radeon: add primary dac adj quirk for R200 board * ALSA: vmaster: Fix slave change notification * hw_random: make buffer usable in scatterlist. * 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 * ARM: 7663/1: perf: fix ARMv7 EVTYPE_MASK to include NSH bit * drm/i915: Don't clobber crtc->fb when queue_flip fails * dm snapshot: add missing module aliases * dm: fix truncated status strings * xen/pci: We don't do multiple MSI's. * cifs: ensure that cifs_get_root() only traverses directories * mwifiex: correct sleep delay counter * ath9k_htc: fix signal strength handling issues * ath9k: fix RSSI dummy marker value * proc connector: reject unprivileged listener bumps * md: raid0: fix error return from create_stripe_zones. * md: fix two bugs when attempting to resize RAID0 array. * ARM: fix scheduling while atomic warning in alignment handling code * ARM: VFP: fix emulation of second VFP instruction * 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 * md: protect against crash upon fsync on ro array * btrfs: Init io_lock after cloning btrfs device struct * Revert "powerpc/eeh: Fix crash when adding a device in a slot with DDW" The following patches had already been applied: * ahci: AHCI-mode SATA patch for Intel Lynx Point DeviceIDs * keys: fix race with concurrent install_user_keyrings() * signal: always clear sa_restorer on execve * crypto: user - fix info leaks in report API * mm/hotplug: correctly add new zone to all other nodes' zone lists * xen-netfront: delay gARP until backend switches to Connected * NLS: improve UTF8 -> UTF16 string conversion routine * tools: hv: Netlink source address validation allows DoS