Oracle Linux 6.1 kernel security, bug fix and enhancement update

[2.6.32-131.0.15.el6]

• [build] disable Werr for external modules (Aristeu Rozanski) [703504]
  [2.6.32-131.0.14.el6]
• [scsi] hpsa: fix reading a write only register causes a hang (Rob Evers) [703262]
• [scsi] mpt2sas: remove the use of writeq, since writeq is not atomic (Tomas Henzl) [701947]
  [2.6.32-131.0.13.el6]
• [scsi] hpsa: fix lost command problem (Tomas Henzl) [700430]
• [scsi] cciss: fix lost command problem (Tomas Henzl) [700430]
• [scsi] ibft: fix oops during boot (Mike Christie) [698737]
  [2.6.32-131.0.12.el6]
• [scsi] beiscsi: update version (Mike Christie) [674340]
• [scsi] be2iscsi: fix chip cleanup (Mike Christie) [674340]
• [scsi] be2iscsi: fix boot hang due to interrupts not getting rearmed (Mike Christie) [674340]
• [scsi] bnx2fc: fix regression due to incorrect setup of em for npiv port (Mike Christie) [700672]
• [ppc] pseries: Use a kmem cache for DTL buffers (Steve Best) [695678]
  [2.6.32-131.0.11.el6]
• [kdump] revert commit 8f4ec27fc to keep crashkernel=auto (Amerigo Wang) [605786]
  [2.6.32-131.0.10.el6]
• [netdrv] cnic: fix hang due to rtnl_lock (Mike Christie) [694874]
• [netdrv] firmware: re-add the recently deleted bnx2x fw 6.2.5.0 (Michal Schmidt) [690470]
• [netdrv] firmware/bnx2x: add 6.2.9.0 fw, remove unused fw (Michal Schmidt) [690470]
• [netdrv] bnx2x, cnic: Disable iSCSI if DCBX negotiation is successful (Michal Schmidt) [690470]
• [netdrv] bnx2x: don’t write dcb/llfc fields in STORM memory (Michal Schmidt) [690470]
• [netdrv] bnx2x: Update firmware to 6.2.9 (Michal Schmidt) [690470]
  [2.6.32-131.0.9.el6]
• [net] limit socket backlog add operation to prevent possible DoS (Jiri Pirko) [694396] {CVE-2010-4251}
• [scsi] mpt2sas: prevent heap overflows and unchecked (Tomas Henzl) [694023] {CVE-2011-1494 CVE-2011-1495}
• [fs] epoll: prevent creating circular epoll structures (Don Howard) [681683] {CVE-2011-1082}
• [mm] Prevent page_fault at do_mm_track_pte+0xc when Stratus dirty page tracking is active (Larry Woodman) [693786]
• [fs] GFS2 causes kernel panic in spectator mode (Steven Whitehouse) [696535]
• [net] bonding: interface doesn’t issue IGMP report on slave interface during failover (Flavio Leitner) [640690]
• [scsi] isci: validate oem parameters early, and fallback (David Milburn) [698016]
• [scsi] isci: fix oem parameter header definition (David Milburn) [698016]
  [2.6.32-131.0.8.el6]
• [scsi] mark bfa fc adapters tech preview (Rob Evers) [698384]
• [virt] Revert pdpte registers are not flushed when PGD entry is changed in x86 PAE mode (Aristeu Rozanski) [691310]
• [i686] nmi watchdog: Enable panic on hardlockup (Don Zickus) [677532]
• [netdrv] Adding Chelsio Firmware for cxgb4 (Neil Horman) [691929]
  [2.6.32-131.0.7.el6]
• [virt] x86: better fix for race between nmi injection and enabling nmi window (Aristeu Rozanski)
• [virt] x86: revert ‘fix race between nmi injection and enabling nmi window’ (Aristeu Rozanski)
  [2.6.32-131.0.6.el6]
• [net] bonding: fix jiffy comparison issues (Andy Gospodarek) [696337]
  [2.6.32-131.0.5.el6]
• [kernel] perf: add script command help (Jiri Olsa) [693050]
• [drm] radeon/kms: make radeon i2c put/get bytes less noisy (Frank Arnold) [693829]
• [drm] radeon/kms: fix hardcoded EDID handling (Frank Arnold) [693829]
• [x86] Revert ‘[x86] perf: P4 PMU - Fix unflagged overflows handling’ (Don Zickus) [688547]
• [x86] perf: let everyone share counters on a P4 machine (Don Zickus) [688547]
• [fs] nfs: Ensure that NFS4 acl requests don’t use slab in skb fraglist (Neil Horman) [682645] {CVE-2011-1090}
• [fs] partitions: Validate map_count in Mac partition tables (Danny Feng) [679286] {CVE-2011-1010}
  [2.6.32-131.0.4.el6]
• [scsi] ibft: search for broadcom specific ibft sign (Mike Christie) [696275]
• [fs] Fix corrupted OSF partition table parsing (Danny Feng) [688025] {CVE-2011-1163}
• [netdrv] ixgbe: DCB, X540 devices do not respond to pause frames (Andy Gospodarek) [694930]
• [netdrv] ixgbe: DCB, misallocated packet buffer size with X540 device (Andy Gospodarek) [694930]
• [netdrv] ixgbe: refactor common start_hw code for 82599 and x54 (Andy Gospodarek) [694930]
• [netdrv] ixgbe: balance free_irq calls with request_irq calls (Andy Gospodarek) [692988]
  [2.6.32-131.0.3.el6]
• [net] sctp: fix the INIT/INIT-ACK chunk length calculation (Thomas Graf) [690743] {CVE-2011-1573}
• [kernel] sched: Fix granularity of task_u/stime() (Jerome Marchand) [690998]
• [pci] Call PCIe _OSC methods earlier (Matthew Garrett) [693974]
• [fs] nfs: use unstable writes for groups of small DIO writes (Jeff Layton) [694309]
• [net] CAN: Use inode instead of kernel address for /proc file (Danny Feng) [664561] {CVE-2010-4565}
• [x86] mce: reject CEs on Westmere EX MCE bank 6 (Prarit Bhargava) [694891]
• [scsi] libfcoe: Incorrect CVL handling for NPIV ports (Mike Christie) [694906]
• [x86] perf: Complain louder about BIOSen corrupting CPU/PMU state and continue (Don Zickus) [694913]
• [fs] inotify: fix double free/corruption of stuct user (Eric Paris) [656832] {CVE-2010-4250}
• [netdrv] netxen: limit skb frags for non tso packet (Chad Dupuis) [695478]
• [fs] nfsd4: fix oops on lock failure (J. Bruce Fields) [696376]
• [netdrv] Return bnx2 firmware files to Makefile (John Feeney) [696365]
• [scsi] be2iscsi: fix be2iscsi rmmod (Mike Christie) [695585]
• [netdrv] qlcnic: limit skb frags for non tso packet (Bob Picco) [695488]
• [md] Cleanup after raid45->raid0 takeover (Dean Nelson) [694106]
• [md] revert ‘Cleanup after raid45->raid0 takeover patch’ (Dean Nelson) [694106]
• [net] bonding: fix incorrect tx queue offset (Andy Gospodarek) [695548] {CVE-2011-1581}
• [netdrv] igb: for 82576 EEPROMs reporting invalid size default to 16kB (Stefan Assmann) [695751]
• [pci] return correct value when writing to the ‘reset’ attribute (Alex Williamson) [690291]
• [kernel] Initalize call_single_queue during boot to handle left over ipi (Neil Horman) [680478]
  [2.6.32-131.0.2.el6]
• [virt] x86: better fix for race between nmi injection and enabling nmi window (Marcelo Tosatti) [684719]
• [virt] x86: revert ‘fix race between nmi injection and enabling nmi window’ (Marcelo Tosatti) [684719]
  [2.6.32-131.0.1.el6]
• [mm] pdpte registers are not flushed when PGD entry is changed in x86 PAE mode (Larry Woodman) [691310]
• [drm] i915: backports from stable to fix some regressions (Dave Airlie) [690865]
• [fs] svcrpc: complete svsk processing on cb receive failure (J. Bruce Fields) [629030]
• [ppc] pseries: fix hang caused by missing spin_unlock in dtl_disable (Steve Best) [694327]
• [ppc] pseries: Disable VPNH feature (Steve Best) [694266]
• [netdrv] bna: Avoid kernel panic in case of FW heartbeat failure (Ivan Vecera) [694115]
• [input] wacom: Move the cintiq initialization down (Peter Hutterer) [693573]
• [input] wacom: specify Cinitq supported tools (Peter Hutterer) [693573]
• [input] wacom: fix pressure in Cintiq 21UX2 (Peter Hutterer) [693573]
• [input] wacom: fix serial number handling on Cintiq 21UX2 (Peter Hutterer) [693573]
• [input] wacom: add Cintiq 21UX2 and Intuos4 WL (Peter Hutterer) [693573]
• [kernel] spec: strip note and comment from ppc64’s vmlinux before checksum is calculated (Aristeu Rozanski) [692515]
• [scsi] fcoe: have fcoe log off and lport destroy before ndo_fcoe_disable (Mike Christie) [691611]
• [scsi] libfc: rec tov value and REC_TOV_CONST units usages is incorrect (Mike Christie) [691611]
• [scsi] libfcoe: fix wrong comment in fcoe_transport_detach (Mike Christie) [691611]
• [scsi] libfcoe: clean up netdev mapping properly when the transport goes away (Mike Christie) [691611]
• [scsi] fcoe: remove unnecessary module state check (Mike Christie) [691611]
• [scsi] fcoe: Remove mutex_trylock/restart_syscall checks (Mike Christie) [691611]
• [scsi] libfcoe: Remove mutex_trylock/restart_syscall checks (Mike Christie) [691611]
• [scsi] fcoe: correct checking for bonding (Mike Christie) [691611]
• [scsi] fcoe: fix broken fcoe interface reset (Mike Christie) [691611]
• [scsi] fcoe: precedence bug in fcoe_filter_frames() (Mike Christie) [691611]
• [scsi] libfcoe: Move FCOE_MTU definition from fcoe.h to libfcoe.h (Mike Christie) [691611]
• [scsi] libfc: remove duplicate ema_list init (Mike Christie) [691611]
• [scsi] fcoe, libfc: initialize EM anchors list and then update npiv EMs (Mike Christie) [691611]
• [scsi] libfc: Fixing a memory leak when destroying an interface (Mike Christie) [691611]
• [scsi] fc: Add GSPN_ID request to header file (Mike Christie) [691611]
• [scsi] hpsa: fix pci_device_id table (Tomas Henzl) [684997]
• [netdrv] ixgbe: only enable WoL for magic packet by default (Andy Gospodarek) [632598]
• [mm] zram: disable zram on ppc64 (Jerome Marchand) [661293]
• [mm] zram: update config file (Jerome Marchand) [661293]
• [mm] zram: initialize device on first read (Jerome Marchand) [661293]
• [mm] zram: fix data corruption issue (Jerome Marchand) [661293]
• [mm] zram: xvmalloc: combine duplicate block delete code (Jerome Marchand) [661293]
• [mm] zram: Return zero’d pages on new reads (Jerome Marchand) [661293]
• [mm] zram: xvmalloc: Close 32byte hole on 64bit CPUs (Jerome Marchand) [661293]
• [mm] zram: xvmalloc: create CONFIG_ZRAM_DEBUG for debug code (Jerome Marchand) [661293]
• [mm] zram: xvmalloc: free bit block insertion optimization (Jerome Marchand) [661293]
• [mm] zram: Prevent overflow in logical block size (Jerome Marchand) [661293]
• [mm] zram: vmalloc: Correct tunings to enable use with 64K pages (Jerome Marchand) [661293]
• [mm] zram: xvmalloc.c: Fix a typo (Jerome Marchand) [661293]
• [mm] zram: Fix sparse warning ‘Using plain integer as NULL pointer’ (Jerome Marchand) [661293]
  [2.6.32-131.el6]
• [tracing] t_start: reset FTRACE_ITER_HASH in case of seek/pread (Jiri Olsa) [631626] {CVE-2010-3079}
• [scsi] scsi_dh_rdac: fix for lun_table update for rdac (Rob Evers) [687878]
• [usb] EHCI: unlink unused QHs when the controller is stopped (Don Zickus) [680987]
• [fs] Revert ‘[fs] sunrpc: Use static const char arrays’ (Steve Dickson) [690754]
• [fs] sunrpc: Propagate errors from xs_bind() through xs_create_sock() (Steve Dickson) [689777]
• [net] netfilter: ipt_CLUSTERIP: fix buffer overflow (Jiri Pirko) [689342]
• [net] ipv6: netfilter: ip6_tables: fix infoleak to userspace (Jiri Pirko) [689351] {CVE-2011-1172}
• [net] netfilter: ip_tables: fix infoleak to userspace (Jiri Pirko) [689334] {CVE-2011-1171}
• [net] netfilter: arp_tables: fix infoleak to userspace (Jiri Pirko) [689325] {CVE-2011-1170}
• [kernel] remove kernel-debuginfo-common requires from perf-debuginfo (Jason Baron) [682012]
• [drm] radeon/kms: check AA resolve registers on r300 + regression fix (Dave Airlie) [680002] {CVE-2011-1016}
• [net] fix ebtables stack infoleak (Eugene Teo) [681323] {CVE-2011-1080}
• [drm] fix unsigned vs signed comparison issue in modeset ctl ioctl (Don Howard) [679928] {CVE-2011-1013}
• [fs] svcrpc: take advantage of tcp autotuning (J. Bruce Fields) [629030]
• [fs] SUNRPC: Don’t wait for full record to receive tcp data (J. Bruce Fields) [629030]
• [net] svcrpc: copy cb reply instead of pages (J. Bruce Fields) [629030]
• [fs] svcrpc: close connection if client sends short packet (J. Bruce Fields) [629030]
• [fs] svcrpc: note network-order types in svc_process_calldir (J. Bruce Fields) [629030]
• [fs] SUNRPC: svc_tcp_recvfrom cleanup (J. Bruce Fields) [629030]
• [fs] SUNRPC: requeue tcp socket less frequently (J. Bruce Fields) [629030]
• [fs] rpc: move sk_bc_xprt to svc_xprt (J. Bruce Fields) [629030]
• [acpi] ACPICA: Truncate I/O addresses to 16 bits for Windows compatibility (Frank Arnold) [593766]
  [2.6.32-130.el6]
• [kernel] kcore: restrict access to the whole memory (Amerigo Wang) [663864]
• [scsi] libsas: flush initial device discovery before completing ->scan_finished (David Milburn) [682265]
• [md] Cleanup after raid45->raid0 takeover (Doug Ledford) [688725]
• [md] partition detection when array becomes active (Doug Ledford) [688725]
• [md] avoid spinlock problem in blk_throtl_exit (Doug Ledford) [679096 688725]
• [md] correctly handle probe of an ‘mdp’ device (Doug Ledford) [688725]
• [md] don’t set_capacity before array is active (Doug Ledford) [688725]
• [md] Fix raid1->raid0 takeover (Doug Ledford) [688725]
• [md] process hangs at wait_barrier after 0->10 takeover (Doug Ledford) [688725]
• [md] md_make_request: don’t touch the bio after calling make_request (Doug Ledford) [688725]
• [md] Don’t allow slot_store while resync/recovery is happening (Doug Ledford) [688725]
• [md] don’t clear curr_resync_completed at end of resync (Doug Ledford) [688725]
• [md] Don’t use remove_and_add_spares to remove failed devices from a read-only array (Doug Ledford) [688725]
• [md] Add raid1->raid0 takeover support (Doug Ledford) [688725]
• [md] Remove the AllReserved flag for component devices (Doug Ledford) [688725]
• [md] don’t abort checking spares as soon as one cannot be added (Doug Ledford) [688725]
• [md] fix the test for finding spares in raid5_start_reshape (Doug Ledford) [688725]
• [md] simplify some ‘if’ conditionals in raid5_start_reshape (Doug Ledford) [688725]
• [md] revert change to raid_disks on failure (Doug Ledford) [688725]
• [md] Fix removal of extra drives when converting RAID6 to RAID5 (Doug Ledford) [688725]
• [md] range check slot number when manually adding a spare (Doug Ledford) [688725]
• [md] raid5: handle manually-added spares in start_reshape (Doug Ledford) [688725]
• [md] fix sync_completed reporting for very large drives (>2TB) (Doug Ledford) [688725]
• [md] allow suspend_lo and suspend_hi to decrease as well as increase (Doug Ledford) [688725]
• [md] Don’t let implementation detail of curr_resync leak out through sysfs (Doug Ledford) [688725]
• [md] separate meta and data devs (Doug Ledford) [688725]
• [md] add new param to_sync_page_io() (Doug Ledford) [688725]
• [md] new param to calc_dev_sboffset (Doug Ledford) [688725]
• [md] Be more careful about clearing flags bit in ->recovery (Doug Ledford) [688725]
• [md] md_stop_writes requires mddev_lock (Doug Ledford) [688725]
• [md] raid5: use sysfs_notify_dirent_safe to avoid NULL pointer (Doug Ledford) [688725]
• [md] Ensure no IO request to get md device before it is properly initialised (Doug Ledford) [688725]
• [md] Fix single printks with multiple KERN_
  s (Doug Ledford) [688725]
• [md] fix regression resulting in delays in clearing bits in a bitmap (Doug Ledford) [688725]
• [md] fix regression with re-adding devices to arrays with no metadata (Doug Ledford) [688725]
• [md] pick some changes from commits to match upstream (Doug Ledford) [688725]
• [md] raid1: add takeover support for raid5->raid1 (Doug Ledford) [688725]
• [md] pick up some percpu annotations that upstream has (Doug Ledford) [688725]
• [md] update includes to match upstream (Doug Ledford) [688725]
• [scsi] isci: fix fragile/conditional isci_host lookups (David Milburn) [691591]
• [scsi] isci: cleanup isci_remote_device[_not]_ready interface (David Milburn) [691591]
• [scsi] isci: Qualify when the host lock is managed for STP/SATA callbacks (David Milburn) [691591]
• [scsi] isci: Fix use of SATA soft reset state machine (David Milburn) [691591]
• [scsi] isci: Free host lock for SATA/STP abort escalation at submission time (David Milburn) [691591]
• [scsi] isci: Properly handle requests in the ‘aborting’ state (David Milburn) [691591]
• [scsi] isci: Remove ‘screaming’ data types (David Milburn) [691591]
• [scsi] isci: remove unused ‘remote_device_started’ (David Milburn) [691591]
• [scsi] isci: namespacecheck cleanups (David Milburn) [691591]
• [scsi] isci: kill some long macros (David Milburn) [691591]
• [scsi] isci: reorder init to cleanup unneeded declarations (David Milburn) [691591]
• [scsi] isci: Remove event_* calls as they are just wrappers (David Milburn) [691591]
• [netdrv] iwlagn: Support new 5000 microcode (Stanislaw Gruszka) [682742]
• [netdrv] iwlwifi: fix dma mappings and skbs leak (Stanislaw Gruszka) [682726]
• [netdrv] iwl3945: remove plcp check (Stanislaw Gruszka) [679002]
• [netdrv] iwlwifi: add {ack,plpc}_check module parameters (Stanislaw Gruszka) [620501]
• [fs] ext4: Fix ext4_quota_write cross block boundary behaviour (Lukas Czerner) [680105]
• [fs] quota: Don’t write quota info in dquot_commit() (Lukas Czerner) [680105]
• [netdrv] be2net: Change f/w command versions for Lancer (Ivan Vecera) [685027]
• [netdrv] be2net: Remove ERR compl workaround for Lancer (Ivan Vecera) [685027]
• [netdrv] be2net: fix to ignore transparent vlan ids wrongly indicated by NIC (Ivan Vecera) [685027]
• [netdrv] be2net: pass proper hdr_size while flashing redboot (Ivan Vecera) [685027]
• [netdrv] be2net: Allow VFs to call be_cmd_reset_function (Ivan Vecera) [685027]
• [netdrv] be2net: pass domain numbers for pmac_add/del functions (Ivan Vecera) [685027]
• [netdrv] be2net: Initialize and cleanup sriov resources only if pci_enable_sriov has succeeded (Ivan Vecera) [685027]
• [netdrv] be2net: Use domain id when be_cmd_if_destroy is called (Ivan Vecera) [685027]
• [netdrv] be2net: While configuring QOS for VF, pass proper domain id (Ivan Vecera) [685027]
• [netdrv] benet: Avoid potential null deref in be_cmd_get_seeprom_data() (Ivan Vecera) [685027]
• [netdrv] benet: fix be_cmd_multicast_set() memcpy bug (Ivan Vecera) [685027]
• [ppc] kdump: Override crash_free_reserved_phys_range to avoid freeing RTAS (Steve Best) [672983]
• [kernel] kdump: Allow shrinking of kdump region to be overridden (Steve Best) [672983]
• [scsi] bnx2fc: Bumped version to 1.0.2 (Mike Christie) [683153]
• [scsi] bnx2fc: Fix kernel panic when deleting NPIV ports (Mike Christie) [683153]
• [scsi] bnx2fc: scsi_dma_unmap() not invoked on IO completions (Mike Christie) [683153]
• [scsi] bnx2fc: host stats show the link speed ‘unknown’ on NIC partitioned interfaces (Mike Christie) [683153]
• [scsi] bnx2fc: IO completion not processed due to missed wakeup (Mike Christie) [683153]
• [scsi] bnx2fc: Bump version to 1.0.1 (Mike Christie) [683153]
• [scsi] bnx2fc: Remove unnecessary module state checks (Mike Christie) [683153]
• [scsi] bnx2fc: Fix MTU issue by using static MTU (Mike Christie) [683153]
• [scsi] bnx2fc: Remove network bonding checking (Mike Christie) [683153]
• [scsi] bnx2fc: Call bnx2fc_return_rqe and bnx2fc_get_next_rqe with tgt lock held (Mike Christie) [683153]
• [scsi] bnx2fc: common free list for cleanup commands (Mike Christie) [683153]
• [scsi] bnx2fc: Remove rtnl_trylock/restart_syscall checks (Mike Christie) [683153]
• [netdrv] cnic: Fix lost interrupt on bnx2x (Mike Christie) [683153]
• [netdrv] cnic: Prevent status block race conditions with hardware (Mike Christie) [683153]
• [kernel] ring-buffer: Use sync sched protection on ring buffer resizing (Jiri Olsa) [676583]
• [kernel] tracing: avoid soft lockup in trace_pipe (Jiri Olsa) [676583]
• [kernel] tracing: Fix a race in function profile (Jiri Olsa) [676583]
• [block] cfq-iosched: Don’t update group weights when on service tree (Vivek Goyal) [689551]
• [block] cfq-iosched: Get rid of on_st flag (Vivek Goyal) [689551]
• [net] tcp_cubic: fix low utilization of CUBIC with HyStart (Thomas Graf) [616985]
• [net] tcp_cubic: make the delay threshold of HyStart less sensitive (Thomas Graf) [616985]
• [net] tcp_cubic: enable high resolution ack time if needed (Thomas Graf) [616985]
• [net] tcp_cubic: fix clock dependency (Thomas Graf) [616985]
• [net] tcp_cubic: make ack train delta value a parameter (Thomas Graf) [616985]
• [net] tcp_cubic: fix comparison of jiffies (Thomas Graf) [616985]
• [net] tcp: fix RTT for quick packets in congestion control (Thomas Graf) [616985]
• [fs] fix GFS2 filesystem hang caused by incorrect lock order (Robert S Peterson) [651584]
• [fs] btrfs: bring us up to date with .38 (Josef Bacik) [684667]
• [ppc] add dynamic dma window support minor updates (Steve Best) [691952]
• [ppc] ptrace: Remove BUG_ON when full register set not available (Steve Best) [678099]
• [ppc] pseries: Disable MSI using new interface if possible (Steve Best) [684961]
• [ppc] kexec: Fix orphaned offline CPUs across kexec (Steve Best) [682875]
• [net] ipsec: Disable granular bundles (Herbert Xu) [631833]
• [scsi] libsas: fix runaway error handler problem (David Milburn) [691527]
• [scsi] mpt2sas: Added customer specific display support (Tomas Henzl) [684841]
• [scsi] Add next gen Dell Powervault controller MD36xxf into RDAC device list (Shyam Iyer) [688979]
• [kernel] perf: Fix task context scheduling (Jiri Olsa) [688065]
• [drm] nouveau: disable acceleration on NVA3/NVA5/NVA8/NVAF by default (Ben Skeggs) [684816]
• [kernel] radix: don’t tag the root if we didn’t tag within our range (Josef Bacik) [681439]
• [block] blk-throttle: Do not use kblockd workqueue for throtl work (Vivek Goyal) [681360]
• [sound] ALSA: HDA hdmi related fixes (Jaroslav Kysela) [671501]
• [pci] Preserve Existing pci sort whitelists for Dell systems (Shyam Iyer) [688954]
• [x86] perf: Add support for AMD family 15h core counters family 15h core counters (Robert Richter) [635671]
• [x86] hpwdt: fix section mismatch warning (Prarit Bhargava) [689837]
• [x86] UV: Correct kABI from upstream (George Beshers) [684957]
• [x86] When cleaning MTRRs, do not fold WP into UC (Prarit Bhargava) [682758]
• [virt] xen-blkfront: handle Xen major numbers other than XENVBD (Andrew Jones) [691339]
• [virt] Fix regression with SMP guests (Zachary Amsden) [681133]
• [netdrv] enic: update to version 2.1.1.13 (Andy Gospodarek) [684865]
• [netdrv] igb: full support for i350 devices (Stefan Assmann) [687932]
• [fs] NFS: Fix a hang/infinite loop in nfs_wb_page() (Steve Dickson) [672305]
• [fs] nfsd: fix auth_domain reference leak on nlm operations (J. Bruce Fields) [690900]
• [fs] svcrpc: ensure cache_check caller sees updated entry (J. Bruce Fields) [690900]
• [fs] svcrpc: take lock on turning entry NEGATIVE in cache_check (J. Bruce Fields) [690900]
• [fs] svcrpc: modifying valid sunrpc cache entries is racy (J. Bruce Fields) [690900]
• [fs] sunrpc: extract some common sunrpc_cache code from nfsd (Steve Dickson) [690900]
• [infiniband] RDMA/cxgb4: Initialization errors can cause crash (Steve Best) [647013]
• [infiniband] RDMA/cxgb4: Don’t change QP state outside EP lock (Steve Best) [647013]
• [infiniband] RDMA/cxgb4: Remove db_drop_task (Steve Best) [647013]
• [infiniband] RDMA/cxgb4: Do CIDX_INC updates every 1/16 CQ depth CQE reaps (Steve Best) [647013]
• [infiniband] RDMA/cxgb4: Dispatch FATAL event on EEH errors (Steve Best) [647013]
• [infiniband] RDMA/cxgb4: Set the correct device physical function for iwarp connections (Steve Best) [647013]
• [infiniband] RDMA/cxgb4: limit MAXBURST EQ context field to 256B (Steve Best) [647013]
• [infiniband] RDMA/cxgb4: Don’t re-init wait object in init/fini paths (Steve Best) [647013]
• [infiniband] RMDA/cxgb4 kfifo changes (Steve Best) [647013]
• [netdrv] cxgb4 driver update (Neil Horman) [647006]
• [tracing] Add unstable sched clock note to the warning (Jiri Olsa) [666264]
• [x86] Reevaluate T-states on CPU hot-add (Matthew Garrett) [673442]
• [scsi] libsas: fix/amend device gone notification in sas_deform_port (David Milburn) [682315]
• [kdump] kexec: move the crashkernel=auto logic into kernel spec file (Amerigo Wang) [605786]