Changelog
linux (3.14.7-1) unstable; urgency=medium
* New upstream stable update:
http://www.kernel.org/pub/linux/kernel/v3.x/ChangeLog-3.14.6
- [mipsel] loongson2_cpufreq: Fix CPU clock rate setting
(regression in 3.14)
- rtmutex: Fix deadlock detector for real
- kernfs: add back missing error check in kernfs_fop_mmap()
(regression in 3.14)
- coredump: fix va_list corruption (regression in 3.11)
- mm: make fixup_user_fault() check the vma access rights too
- serial: 8250: Fix thread unsafe __dma_tx_complete function
- 8250_core: Fix unwanted TX chars write
- iwlwifi: 7000: bump API to 9
- timer: Prevent overflow in apply_slack
- cfg80211: free sme on connection failures (regression in 3.11)
- cfg80211: add cfg80211_sched_scan_stopped_rtnl (regression in 3.14)
- mac80211: fix nested rtnl locking on ieee80211_reconfig
(regression in 3.14)
- mm, thp: close race between mremap() and split_huge_page()
- [x86] mm, hugetlb: Add missing TLB page invalidation for hugetlb_cow()
- hwpoison, hugetlb: lock_page/unlock_page does not match for handling a
free hugepage
- iwlwifi: mvm: delay enabling smart FIFO until after beacon RX
(regression in 3.14)
- aio: fix potential leak in aio_run_iocb().
- Revert "hwmon: (coretemp) Refine TjMax detection"
- hrtimer: Prevent remote enqueue of leftmost timers
- hrtimer: Set expiry time before switch_hrtimer_base()
- dm verity: fix biovecs hash calculation regression (regression in 3.14)
- dm cache: fix writethrough mode quiescing in cache_map
(regression in 3.13)
- md/raid10: call wait_barrier() for each request submitted.
(regression in 3.14)
- PNP / ACPI: Do not return errors if _DIS or _SRS are not present
(regression in 3.14)
- ACPI / EC: Process rather than discard events in acpi_ec_clear
(regression in 3.13.7, 3.14)
- irqchip: armada-370-xp: fix invalid cast of signed value into unsigned
variable (regression in 3.13)
- irqchip: armada-370-xp: implement the ->check_device() msi_chip
operation (regression in 3.13)
- irqchip: armada-370-xp: Fix releasing of MSIs (regression in 3.13)
- [x86] drm/i915: Allow user modes to exceed DVI 165MHz limit
(regression in 3.14)
- [x86] drm/i915: Don't check gmch state on inherited configs
(regression in 3.13?)
- [x86] drm/i915: Don't WARN nor handle unexpected hpd interrupts on gmch
platforms (regression in 3.13)
- [x86] drm/radeon: fix runpm handling on APUs (v4) (regression in 3.13)
- drm/radeon: disable mclk dpm on R7 260X (regression in 3.14)
- drm/radeon: add support for newer mc ucode on SI (v2)
- drm/radeon: add support for newer mc ucode on CI (v2)
- drm/radeon: re-enable mclk dpm on R7 260X asics
- drm/radeon/uvd: use lower clocks on old UVD to boot v2
(regression in 3.13)
- drm/radeon: check buffer relocation offset
- USB: Nokia 305 should be treated as unusual dev
- USB: Nokia 5300 should be treated as unusual dev
- Revert "Bluetooth: Enable autosuspend for Intel Bluetooth device"
(regression in 3.14)
- posix_acl: handle NULL ACL in posix_acl_equiv_mode
- fs/affs/super.c: bugfix / double free (regression in 3.14)
- [armel/orion5x] fix target ID for crypto SRAM window
(regression in 3.12)
- [armel/kirkwood]: dts: fix mislocated pcie-controller nodes
(regression in 3.12)
- [armhf/armmp-lpae] 8012/1: kdump: Avoid overflow when converting pfn to
physaddr
- drm/nouveau: fix another lock unbalance in nouveau_crtc_page_flip
(regression in 3.11)
- drm/i915/vlv: reset VLV media force wake request register
(regression in 3.14?)
- i40e: potential array underflow in i40e_vc_process_vf_msg()
- igb: Fix Null-pointer dereference in igb_reset_q_vector
(regression in 3.14)
- igb: Unset IGB_FLAG_HAS_MSIX-flag when falling back to msi-only
(regression in 3.14)
- leds: leds-pwm: properly clean up after probe failure
- device_cgroup: rework device access check and exception checking
- device_cgroup: check if exception removal is allowed
- media: media-device: fix infoleak in ioctl media_enum_entities()
(CVE-2014-1739)
- Input: Add INPUT_PROP_TOPBUTTONPAD device property
- Input: synaptics - report INPUT_PROP_TOPBUTTONPAD property
- e1000e: Fix no connectivity when driver loaded with cable out
(regression in 3.12)
- autofs: fix lockref lookup
- vfs: fix races between __d_instantiate() and checks of dentry flags
- ALSA: hda - hdmi: Set converter channel count even without sink
(regression in 3.13)
- NFSd: Move default initialisers from create_client() to alloc_client()
- NFSd: call rpc_destroy_wait_queue() from free_client()
- NFSD: Call ->set_acl with a NULL ACL structure if no entries
- nfsd4: remove lockowner when removing lock stateid
- workqueue: fix bugs in wq_update_unbound_numa() failure path
- workqueue: fix a possible race condition between rescuer and pwq-release
- [arm] mvebu: mvebu-soc-id: add missing clk_put() call
(regression in 3.14)
- [arm] mvebu: mvebu-soc-id: keep clock enabled if PCIe unit is enabled
(regression in 3.14)
- ASoC: dapm: Skip CODEC<->CODEC links in connect_dai_link_widgets()
(regression in 3.14)
- [hppa] ratelimit userspace segfault printing
- [amd64] modify_ldt: Make support for 16-bit segments a runtime option
- sysfs: make sure read buffer is zeroed (possible regression in 3.13)
- Target/iser: Fix wrong connection requests list addition
- Target/iser: Fix iscsit_accept_np and rdma_cm racy flow
- iscsi-target: Change BUG_ON to REJECT in iscsit_process_nop_out
(regression in 3.11)
- target: fix memory leak on XCOPY
- [x86] drm/i915: Disable self-refresh for untiled fbs on i915gm
(regression in 3.14)
- [x86] drm/i915: move power domain init earlier during system resume
(regression in 3.14?)
- [x86] drm/i915: Fix unsafe loop iteration over vma whilst unbinding them
(regression in 3.12)
- iwlwifi: mvm: BT Coex - fix Look Up Table (regression in 3.13)
- PCI: Wrong register used to check pending traffic (regression in 3.14)
- dm crypt: fix cpu hotplug crash by removing per-cpu structure
- dm thin: allow metadata commit if pool is in PM_OUT_OF_DATA_SPACE mode
(regression in 3.14)
- dm thin: add timeout to stop out-of-data-space mode holding IO forever
- dmaengine: fix dmaengine_unmap failure
- dma: mv_xor: Flush descriptors before activating a channel
- tcm_fc: Fix free-after-use regression in ft_free_cmd
(regression in 3.13)
- ACPICA: Tables: Restore old behavor to favor 32-bit FADT addresses.
(regression in 3.14)
- ACPI: Revert "ACPI: Remove CONFIG_ACPI_PROCFS_POWER and cm_sbsc.c"
(regression in 3.13)
- ACPI: Revert "ACPI / Battery: Remove battery's proc directory"
(regression in 3.13)
- [x86] ACPI / video: Add use_native_backlight quirks for more systems
- ACPI: Revert "ACPI / AC: convert ACPI ac driver to platform bus"
(regression in 3.13)
- [x86] ACPI / TPM: Fix resume regression on Chromebooks
(regression in 3.14)
- i2c: s3c2410: resume race fix
- [x86] intel_pstate: Set turbo VID for BayTrail
- [s390] crypto: fix aes,des ctr mode concurrency finding.
- clk: Fix double free due to devm_clk_register()
- clk: Fix slab corruption in clk_unregister()
- [powerpc] powernv: Reset root port in firmware (regression in 3.14)
- [powerpc] irq work racing with timer interrupt can result in timer
interrupt hang (regression in 3.14)
- [powerpc] kexec: Fix "Processor X is stuck" issue during kexec from ST
mode (regression in 3.13)
- spi: core: Ignore unsupported Dual/Quad Transfer Mode bits
(regression in 3.12)
- libceph: fix corruption when using page_count 0 page in rbd
- media: V4L2: ov7670: fix a wrong index, potentially Oopsing the kernel
from user-space
http://www.kernel.org/pub/linux/kernel/v3.x/ChangeLog-3.14.7
- perf: Limit perf_event_attr::sample_period to 63 bits
- perf: Fix race in removing an event
- SCSI: scsi_transport_sas: move bsg destructor into sas_rphy_remove
(regression in 3.13)
- [arm] 8051/1: put_user: fix possible data corruption in put_user
- [arm] 8064/1: fix v7-M signal return (regression in 3.11)
- cpufreq: remove race while accessing cur_policy
- firewire: revert to 4 GB RDMA, fix protocols using Memory Space
(regression in 3.14)
- [mips] Fix typo when reporting cache and ftlb errors for ImgTec cores
(regression in 3.14)
- dm cache: always split discards on cache block boundaries
- virtio_blk: fix race between start and stop queue
- sched/deadline: Restrict user params max value to 2^63 ns
- sched/dl: Fix race in dl_task_timer()
- drm/radeon: avoid crash if VM command submission isn't available
- [x86] drm/i915: Only copy back the modified fields to userspace from
execbuffer
- drm/radeon/dpm: resume fixes for some systems (regression in 3.14)
- libata: Blacklist queued trim for Crucial M500 (regression in 3.14.4)
- md: always set MD_RECOVERY_INTR when aborting a reshape or other
"resync".
- md: always set MD_RECOVERY_INTR when interrupting a reshape thread.
(regression in 3.12.9, 3.13)
- xhci: delete endpoints from bandwidth list before freeing whole device
- staging: comedi: ni_daq_700: add mux settling delay
- staging: r8192e_pci: fix htons error (regression in 3.14)
- ALSA: hda/analog - Fix silent output on ASUS A8JN (regression in 3.12)
- USB: io_ti: fix firmware download on big-endian machines (part 2)
- usb: pci-quirks: Prevent Sony VAIO t-series from switching usb ports
(regression in 3.12)
- percpu-refcount: fix usage of this_cpu_ops
- [x86] intel_pstate: Remove C0 tracking (regression in 3.14)
- [x86] intel_pstate: Correct rounding in busy calculation
(regression in 3.14)
- [x86] intel_pstate: add sample time scaling
- [x86] intel_pstate: Improve initial busy calculation
- mm: add !pte_present() check on existing hugetlb_entry callbacks
(CVE-2014-3940)
- mm: rmap: fix use-after-free in __put_anon_vma
- iser-target: Add missing target_put_sess_cmd for ImmedateData failure
- iscsi-target: Fix wrong buffer / buffer overrun in
iscsi_change_param_value()
- target: Fix alua_access_state attribute OOPs for un-configured devices
- netfilter: Fix potential use after free in ip6_route_me_harder()
- netfilter: nfnetlink: Fix use after free when it fails to process batch
- [x86] iommu/vt-d: Fix missing IOTLB flush in intel_iommu_unmap()
[ Ian Campbell ]
* [armhf] Enable VIRTIO_BALLOON and VIRTIO_PCI (Closes: #750742)
* [arm64] Update modules included in installer udebs.
* Include virtio_mmio in virtio-modules udeb when available.
[ Aurelien Jarno ]
* topconfig: Enable modular HW_RANDOM.
* [kernelarch-powerpc] Remove HW_RANDOM.
* [kernelarch-x86] Remove HW_RANDOM.
* [mips/4kc-malta] Remove HW_RANDOM.
* [mips/5kc-malta] Remove HW_RANDOM.
* [mips/octeon] Remove HW_RANDOM.
* [arm64] Enable modular RTC_DRV_PL031.
[ Ben Hutchings ]
* [ppc64el] Add kernel image, thanks to Mauricio Faria de Oliveira:
- Split common/big-endian powerpc64 options
- Add little-endian powerpc64 options
- Temporarily disable zImage
- powerpc/powernv: Add calls to support little endian host
- Add 'ppc64le' (uname output) to bug/include-model script
- udeb: Add packages based on ppc64 configuration
* netfilter: ipv4: defrag: set local_df flag on defragmented skb
(regression in 3.14.5)
* [mips] seccomp: Check system calls whenever seccomp is enabled,
even if audit and trace are disabled (Closes: #751417)
* auditsc: audit_krule mask accesses need bounds checking (CVE-2014-3917)
* fs,userns: Change inode_capable to capable_wrt_inode_uidgid
(CVE-2014-4014)
* SCSI: Fix spurious request sense in error handling (regression in 3.14)
* PCI/MSI: Fix memory leak in free_msi_irqs() (regression in 3.14)
* [rt] hrtimer: Disable MISSED_TIMER_OFFSETS_HIST as it will currently
result in a panic
-- Ben Hutchings <email address hidden> Mon, 16 Jun 2014 09:51:49 +0100
