)]}'
{
  "log": [
    {
      "commit": "d1db0eea852497762cab43b905b879dfcd3b8987",
      "tree": "27bd827450c836cc2d6066dbf0a6110062feea2b",
      "parents": [
        "ec6931b281797b69e6cf109f9cc94d5a2bf994e0"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Sun Apr 27 19:29:27 2014 -0700"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Sun Apr 27 19:29:27 2014 -0700"
      },
      "message": "Linux 3.15-rc3\n"
    },
    {
      "commit": "ec6931b281797b69e6cf109f9cc94d5a2bf994e0",
      "tree": "9e8ab9ff709939a2ca13a7a5556b436689e2908b",
      "parents": [
        "ac6c9e2bed093c4b60e313674fb7aec4f264c3d4"
      ],
      "author": {
        "name": "Will Deacon",
        "email": "will.deacon@arm.com",
        "time": "Wed Apr 23 17:52:52 2014 +0100"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Sun Apr 27 15:20:05 2014 -0700"
      },
      "message": "word-at-a-time: avoid undefined behaviour in zero_bytemask macro\n\nThe asm-generic, big-endian version of zero_bytemask creates a mask of\nbytes preceding the first zero-byte by left shifting ~0ul based on the\nposition of the first zero byte.\n\nUnfortunately, if the first (top) byte is zero, the output of\nprep_zero_mask has only the top bit set, resulting in undefined C\nbehaviour as we shift left by an amount equal to the width of the type.\nAs it happens, GCC doesn\u0027t manage to spot this through the call to fls(),\nbut the issue remains if architectures choose to implement their shift\ninstructions differently.\n\nAn example would be arch/arm/ (AArch32), where LSL Rd, Rn, #32 results\nin Rd \u003d\u003d 0x0, whilst on arch/arm64 (AArch64) LSL Xd, Xn, #64 results in\nXd \u003d\u003d Xn.\n\nRather than check explicitly for the problematic shift, this patch adds\nan extra shift by 1, replacing fls with __fls. Since zero_bytemask is\nnever called with a zero argument (has_zero() is used to check the data\nfirst), we don\u0027t need to worry about calling __fls(0), which is\nundefined.\n\nCc: \u003cstable@vger.kernel.org\u003e\nCc: Victor Kamensky \u003cvictor.kamensky@linaro.org\u003e\nSigned-off-by: Will Deacon \u003cwill.deacon@arm.com\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n"
    },
    {
      "commit": "ac6c9e2bed093c4b60e313674fb7aec4f264c3d4",
      "tree": "6eccadffcc04a7b66021ab3bd114cafcb52181c6",
      "parents": [
        "33c0022f0e687b0161a9bb84a5671df932551e3a",
        "1cf35d47712dd5dc4d62c6ce984f04ac6eab0408"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Sun Apr 27 15:08:12 2014 -0700"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Sun Apr 27 15:08:12 2014 -0700"
      },
      "message": "Merge branch \u0027safe-dirty-tlb-flush\u0027\n\nThis merges the patch to fix possible loss of dirty bit on munmap() or\nmadvice(DONTNEED).  If there are concurrent writers on other CPU\u0027s that\nhave the unmapped/unneeded page in their TLBs, their writes to the page\ncould possibly get lost if a third CPU raced with the TLB flush and did\na page_mkclean() before the page was fully written.\n\nAdmittedly, if you unmap() or madvice(DONTNEED) an area _while_ another\nthread is still busy writing to it, you deserve all the lost writes you\ncould get.  But we kernel people hold ourselves to higher quality\nstandards than \"crazy people deserve to lose\", because, well, we\u0027ve seen\npeople do all kinds of crazy things.\n\nSo let\u0027s get it right, just because we can, and we don\u0027t have to worry\nabout it.\n\n* safe-dirty-tlb-flush:\n  mm: split \u0027tlb_flush_mmu()\u0027 into tlb flushing and memory freeing parts\n"
    },
    {
      "commit": "33c0022f0e687b0161a9bb84a5671df932551e3a",
      "tree": "af5a871d8685c0baca9c4b8fcf31cb9f5454374a",
      "parents": [
        "2b9d1c050d291693f257c98605028325f8146c84",
        "cfd4a535b68faf651b238586011f5bae128391c4"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Sun Apr 27 13:26:28 2014 -0700"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Sun Apr 27 13:26:28 2014 -0700"
      },
      "message": "Merge branch \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/mason/linux-btrfs\n\nPull btrfs fixes from Chris Mason.\n\n* \u0027for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/mason/linux-btrfs:\n  Btrfs: limit the path size in send to PATH_MAX\n  Btrfs: correctly set profile flags on seqlock retry\n  Btrfs: use correct key when repeating search for extent item\n  Btrfs: fix inode caching vs tree log\n  Btrfs: fix possible memory leaks in open_ctree()\n  Btrfs: avoid triggering bug_on() when we fail to start inode caching task\n  Btrfs: move btrfs_{set,clear}_and_info() to ctree.h\n  btrfs: replace error code from btrfs_drop_extents\n  btrfs: Change the hole range to a more accurate value.\n  btrfs: fix use-after-free in mount_subvol()\n"
    },
    {
      "commit": "2b9d1c050d291693f257c98605028325f8146c84",
      "tree": "756d061271ae942141aaecd8656b035f562f4906",
      "parents": [
        "afa3cad746cc0fb547f1ab8d881db445e237840f",
        "d93003e8e4e1fbbc8a06ec561a63f5aa105a4c45"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Sun Apr 27 12:55:04 2014 -0700"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Sun Apr 27 12:55:04 2014 -0700"
      },
      "message": "Merge branch \u0027fixes\u0027 of git://ftp.arm.linux.org.uk/~rmk/linux-arm\n\nPull arm fixes from Russell King:\n \"A number of fixes for the PJ4/iwmmxt changes which arm-soc forced me\n  to take during the merge window.  This stuff should have been better\n  tested and sorted out *before* the merge window\"\n\n* \u0027fixes\u0027 of git://ftp.arm.linux.org.uk/~rmk/linux-arm:\n  ARM: 8042/1: iwmmxt: allow to build iWMMXt on Marvell PJ4B\n  ARM: 8041/1: pj4: fix cpu_is_pj4 check\n  ARM: 8040/1: pj4: properly detect existence of iWMMXt coprocessor\n  ARM: 8039/1: pj4: enable iWMMXt only if CONFIG_IWMMXT is set\n  ARM: 8038/1: iwmmxt: explicitly check for supported architectures\n"
    },
    {
      "commit": "afa3cad746cc0fb547f1ab8d881db445e237840f",
      "tree": "842bd8c0955397531a96bdffbb92629c775248d7",
      "parents": [
        "d9e9e8e2fe832180f5c8f659a63def2e8fcaea4a",
        "bc3ee18a7a57243721ecfd879319e3d2e882f289"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Sun Apr 27 12:54:05 2014 -0700"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Sun Apr 27 12:54:05 2014 -0700"
      },
      "message": "Merge tag \u0027arm64-fixes\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/arm64/linux\n\nPull arm64 fixes from Catalin Marinas:\n - compat renameat2 syscall wiring and __NR_compat_syscalls fix\n - TLB fix for transparent huge pages following switch to generic\n   mmu_gather\n - spinlock initialisation for init_mm\u0027s context\n - move of_clk_init() earlier\n - Kconfig duplicate entry fix\n\n* tag \u0027arm64-fixes\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/arm64/linux:\n  arm64: init: Move of_clk_init to time_init\n  arm64: initialize spinlock for init_mm\u0027s context\n  arm64: debug: remove noisy, pointless warning\n  arm64: mm: Add THP TLB entries to general mmu_gather\n  arm64: add renameat2 compat syscall\n  ARM64: Remove duplicated Kconfig entry for \"kernel/power/Kconfig\"\n  arm64: __NR_compat_syscalls fix\n"
    },
    {
      "commit": "d9e9e8e2fe832180f5c8f659a63def2e8fcaea4a",
      "tree": "bac04f7aeb4576dc27a6d52c0cbf8fdfa81ecb6f",
      "parents": [
        "a8d706986c5ee65354d8ddd88fe2dadfd2184991",
        "8db6e5104b77de5d0b7002b95069da0992a34be9"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Sun Apr 27 11:21:03 2014 -0700"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Sun Apr 27 11:21:03 2014 -0700"
      },
      "message": "Merge branch \u0027irq-urgent-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip\n\nPull irq fixes from Thomas Gleixner:\n \"A slighlty large fix for a subtle issue in the CPU hotplug code of\n  certain ARM SoCs, where the not yet online cpu needs to setup the cpu\n  local timer and needs to set the interrupt affinity to itself.\n  Setting interrupt affinity to a not online cpu is prohibited and\n  therefor the timer interrupt ends up on the wrong cpu, which leads to\n  nasty complications.\n\n  The SoC folks tried to hack around that in the SoC code in some more\n  than nasty ways.  The proper solution is to have a way to enforce the\n  affinity setting to a not online cpu.  The core patch to the genirq\n  code provides that facility and the follow up patches make use of it\n  in the GIC interrupt controller and the exynos timer driver.\n\n  The change to the core code has no implications to existing users,\n  except for the rename of the locked function and therefor the\n  necessary fixup in mips/cavium.  Aside of that, no runtime impact is\n  possible, as none of the existing interrupt chips implements anything\n  which depends on the force argument of the irq_set_affinity()\n  callback\"\n\n* \u0027irq-urgent-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:\n  clocksource: Exynos_mct: Register clock event after request_irq()\n  clocksource: Exynos_mct: Use irq_force_affinity() in cpu bringup\n  irqchip: Gic: Support forced affinity setting\n  genirq: Allow forcing cpu affinity of interrupts\n"
    },
    {
      "commit": "a8d706986c5ee65354d8ddd88fe2dadfd2184991",
      "tree": "274e9e68933010e179673e35d635c4acef51eb56",
      "parents": [
        "d0c15ad760c869f3d676b37c2989baa1566d31f1",
        "7deb39ed8d9494ea541bcaa69b56036a94f79dc2"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Sun Apr 27 10:39:09 2014 -0700"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Sun Apr 27 10:39:09 2014 -0700"
      },
      "message": "Merge tag \u0027tty-3.15-rc3\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/tty\n\nPull tty/serial fixes from Greg KH:\n \"Here are a few tty/serial fixes for 3.15-rc3 that resolve a number of\n  reported issues in the 8250 and samsung serial drivers, as well as a\n  character loss fix for the tty core that was caused by the lock\n  removal patches a release ago\"\n\n* tag \u0027tty-3.15-rc3\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/tty:\n  serial_core: fix uart PORT_UNKNOWN handling\n  serial: samsung: Change barrier() to cpu_relax() in console output\n  serial: samsung: don\u0027t check config for every character\n  serial: samsung: Use the passed in \"port\", fixing kgdb w/ no console\n  serial: 8250: Fix thread unsafe __dma_tx_complete function\n  8250_core: Fix unwanted TX chars write\n  tty: Fix race condition between __tty_buffer_request_room and flush_to_ldisc\n"
    },
    {
      "commit": "d0c15ad760c869f3d676b37c2989baa1566d31f1",
      "tree": "db1654d6324d0519fd39b91fb172df2654ee2c93",
      "parents": [
        "005fbcd03458b6c0044f65c1fe8d8a68d726a321",
        "2704f807f9498054b8153002bafa3e818079e9a5"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Sun Apr 27 10:34:29 2014 -0700"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Sun Apr 27 10:34:29 2014 -0700"
      },
      "message": "Merge tag \u0027staging-3.15-rc3\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/staging\n\nPull staging / IIO driver fixes from Greg KH:\n \"Here are some small staging and IIO driver fixes for 3.15-rc3.\n\n  Nothing major at all, just some assorted issues that people have\n  reported\"\n\n* tag \u0027staging-3.15-rc3\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/staging:\n  staging: comedi: usbdux: bug fix for accessing \u0027ao_chanlist\u0027 in private data\n  iio: adc: mxs-lradc: fix warning when buidling on avr32\n  iio: cm36651: Fix i2c client leak and possible NULL pointer dereference\n  iio: querying buffer scan_mask should return 0/1\n  staging:iio:ad2s1200 fix a missing break\n  iio: adc: at91_adc: correct default shtim value\n  ARM: at91: at91sam9260: change at91_adc name\n  ARM: at91: at91sam9g45: change at91_adc name\n  iio: cm32181: Fix read integration time function\n  iio: adc: at91_adc: Repair broken platform_data support\n"
    },
    {
      "commit": "005fbcd03458b6c0044f65c1fe8d8a68d726a321",
      "tree": "60265e75c50f3acf802e195363f44b8a10acdf80",
      "parents": [
        "fefb82756e9fd89d0c0d5983e056df81d51eb075",
        "0c8c77d35582c3f7989f1316368da5ae7f14ad4b"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Sun Apr 27 10:28:34 2014 -0700"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Sun Apr 27 10:28:34 2014 -0700"
      },
      "message": "Merge tag \u0027driver-core-3.15-rc3\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/driver-core\n\nPull driver core fixes from Greg KH:\n \"Here are some kernfs fixes for 3.15-rc3 that resolve some reported\n  problems.  Nothing huge, but all needed\"\n\n* tag \u0027driver-core-3.15-rc3\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/driver-core:\n  s390/ccwgroup: Fix memory corruption\n  kernfs: add back missing error check in kernfs_fop_mmap()\n  kernfs: fix a subdir count leak\n"
    },
    {
      "commit": "fefb82756e9fd89d0c0d5983e056df81d51eb075",
      "tree": "f68d8d894d915fa64502bad8a0b83c811e2ff2e2",
      "parents": [
        "e9dba837640d960f56bef22ff08611955ff8a5b4",
        "34f972d6156fe9eea2ab7bb418c71f9d1d5c8e7b"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Sun Apr 27 10:24:17 2014 -0700"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Sun Apr 27 10:24:17 2014 -0700"
      },
      "message": "Merge tag \u0027usb-3.15-rc3\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb\n\nPull USB fixes from Greg KH:\n \"Here are a number of USB fixes for 3.15-rc3.  The majority are gadget\n  fixes, as we didn\u0027t get any of those in for 3.15-rc2.  The others are\n  all over the place, and there\u0027s a number of new device id addtions as\n  well.\"\n\n* tag \u0027usb-3.15-rc3\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb: (35 commits)\n  usb: option: add and update a number of CMOTech devices\n  usb: option: add Alcatel L800MA\n  usb: option: add Olivetti Olicard 500\n  usb: qcserial: add Sierra Wireless MC7305/MC7355\n  usb: qcserial: add Sierra Wireless MC73xx\n  usb: qcserial: add Sierra Wireless EM7355\n  USB: io_ti: fix firmware download on big-endian machines\n  usb/xhci: fix compilation warning when !CONFIG_PCI \u0026\u0026 !CONFIG_PM\n  xhci: extend quirk for Renesas cards\n  xhci: Switch Intel Lynx Point ports to EHCI on shutdown.\n  usb: xhci: Prefer endpoint context dequeue pointer over stopped_trb\n  phy: core: make NULL a valid phy reference if !CONFIG_GENERIC_PHY\n  phy: fix kernel oops in phy_lookup()\n  phy: restore OMAP_CONTROL_PHY dependencies\n  phy: exynos: fix building as a module\n  USB: serial: fix sysfs-attribute removal deadlock\n  usb: wusbcore: fix panic in wusbhc_chid_set\n  usb: wusbcore: convert nested lock to use spin_lock instead of spin_lock_irq\n  uwb: don\u0027t call spin_unlock_irq in a USB completion handler\n  usb: chipidea: coordinate usb phy initialization for different phy type\n  ...\n"
    },
    {
      "commit": "e9dba837640d960f56bef22ff08611955ff8a5b4",
      "tree": "ebeeddd5aa51d6f9f48f3a15ca3690c56267059d",
      "parents": [
        "9a60ee117bbeaf2fb9a02ea80a6bdbc2811ca4d2",
        "d4c9c8a09cff85f69b2c224ca7dbeb42431da2b3"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Sun Apr 27 10:19:06 2014 -0700"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Sun Apr 27 10:19:06 2014 -0700"
      },
      "message": "Merge tag \u0027pm+acpi-3.15-rc3\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm\n\nPull ACPI and power management fixes from Rafael Wysocki:\n \"These include a fix for a recent ACPI regression related to device\n  notifications, intel_idle fix related to IvyTown support, fix for a\n  buffer size issue in ACPICA, PM core fix related to the \"freeze\" sleep\n  state, four fixes for various types of breakage in cpufreq drivers, a\n  PNP workaround for a wrong memory region size in ACPI tables, and a\n  fix and cleanup for the ACPI tools Makefile.\n\n  Specifics:\n\n   - Fix for broken ACPI notifications on some systems caused by a\n     recent ACPI hotplug commit that blocked the propagation of unknown\n     type notifications to device drivers inadvertently.\n\n   - intel_idle fix to make the IvyTown C-states handling (added\n     recently) work as intended which now is broken due to missing\n     braces.  From Christoph Jaeger.\n\n   - ACPICA fix to make it allocate buffers of the right sizes for the\n     Generic Serial Bus operation region access.  From Lv Zheng.\n\n   - PM core fix unblocking cpuidle before entering the \"freeze\" sleep\n     state which causes that state to be able to actually save more\n     energy than runtime idle.\n\n   - Configuration and build fixes for the highbank and powernv cpufreq\n     drivers from Kefeng Wang and Srivatsa S Bhat.\n\n   - Coccinelle warning fix related to error pointers for the unicore32\n     cpufreq driver from Duan Jiong.\n\n   - Integer overflow fix for the ppc-corenet cpufreq driver from Geert\n     Uytterhoeven.\n\n   - Workaround for BIOSes that don\u0027t report the entire Intel MCH area\n     in their ACPI tables from Bjorn Helgaas.\n\n   - ACPI tools Makefile fix and cleanup from Thomas Renninger\"\n\n* tag \u0027pm+acpi-3.15-rc3\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm:\n  ACPI / notify: Do not block unknown type notifications in root handler\n  PNP: Work around BIOS defects in Intel MCH area reporting\n  cpufreq: highbank: fix ARM_HIGHBANK_CPUFREQ dependency warning\n  cpufreq: ppc: Fix integer overflow in expression\n  cpufreq, powernv: Fix build failure on UP\n  cpufreq: unicore32: replace IS_ERR and PTR_ERR with PTR_ERR_OR_ZERO\n  PM / suspend: Make cpuidle work in the \"freeze\" state\n  intel_idle: fix IVT idle state table setting\n  ACPICA: Fix buffer allocation issue for generic_serial_bus region accesses.\n  tools/power/acpi: Minor bugfixes\n"
    },
    {
      "commit": "cfd4a535b68faf651b238586011f5bae128391c4",
      "tree": "03380f86411c67a7ba84d3af697dcdfd01b9f66f",
      "parents": [
        "f8213bdc89719bad895a02c62c4a85066ff76720"
      ],
      "author": {
        "name": "Chris Mason",
        "email": "clm@fb.com",
        "time": "Sat Apr 26 05:02:03 2014 -0700"
      },
      "committer": {
        "name": "Chris Mason",
        "email": "clm@fb.com",
        "time": "Sat Apr 26 05:02:03 2014 -0700"
      },
      "message": "Btrfs: limit the path size in send to PATH_MAX\n\nfs_path_ensure_buf is used to make sure our path buffers for\nsend are big enough for the path names as we construct them.\nThe buffer size is limited to 32K by the length field in\nthe struct.\n\nBut bugs in the path construction can end up trying to build\na huge buffer, and we\u0027ll do invalid memmmoves when the\nbuffer length field wraps.\n\nThis patch is step one, preventing the overflows.\n\nSigned-off-by: Chris Mason \u003cclm@fb.com\u003e\n"
    },
    {
      "commit": "1cf35d47712dd5dc4d62c6ce984f04ac6eab0408",
      "tree": "f00857df7a2eec9520c1a950a0f9ae16cdfc4627",
      "parents": [
        "9a60ee117bbeaf2fb9a02ea80a6bdbc2811ca4d2"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Fri Apr 25 16:05:40 2014 -0700"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Fri Apr 25 16:05:40 2014 -0700"
      },
      "message": "mm: split \u0027tlb_flush_mmu()\u0027 into tlb flushing and memory freeing parts\n\nThe mmu-gather operation \u0027tlb_flush_mmu()\u0027 has done two things: the\nactual tlb flush operation, and the batched freeing of the pages that\nthe TLB entries pointed at.\n\nThis splits the operation into separate phases, so that the forced\nbatched flushing done by zap_pte_range() can now do the actual TLB flush\nwhile still holding the page table lock, but delay the batched freeing\nof all the pages to after the lock has been dropped.\n\nThis in turn allows us to avoid a race condition between\nset_page_dirty() (as called by zap_pte_range() when it finds a dirty\nshared memory pte) and page_mkclean(): because we now flush all the\ndirty page data from the TLB\u0027s while holding the pte lock,\npage_mkclean() will be held up walking the (recently cleaned) page\ntables until after the TLB entries have been flushed from all CPU\u0027s.\n\nReported-by: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\nTested-by: Dave Hansen \u003cdave.hansen@intel.com\u003e\nAcked-by: Hugh Dickins \u003chughd@google.com\u003e\nCc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\nCc: Russell King - ARM Linux \u003clinux@arm.linux.org.uk\u003e\nCc: Tony Luck \u003ctony.luck@intel.com\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n"
    },
    {
      "commit": "d4c9c8a09cff85f69b2c224ca7dbeb42431da2b3",
      "tree": "7bcb901a64d9521c1f235b80972edbe1094fb95b",
      "parents": [
        "e38ed294298089362a224a3f601efe6a49d1a763",
        "cb171f7abb9a1a250fb41d088b81799f75bd1357",
        "f66abe92ce498672ac44d550f51a526597c731c4"
      ],
      "author": {
        "name": "Rafael J. Wysocki",
        "email": "rafael.j.wysocki@intel.com",
        "time": "Sat Apr 26 00:40:25 2014 +0200"
      },
      "committer": {
        "name": "Rafael J. Wysocki",
        "email": "rafael.j.wysocki@intel.com",
        "time": "Sat Apr 26 00:40:25 2014 +0200"
      },
      "message": "Merge branches \u0027pnp\u0027 and \u0027acpi-hotplug\u0027\n\n* pnp:\n  PNP: Work around BIOS defects in Intel MCH area reporting\n\n* acpi-hotplug:\n  ACPI / notify: Do not block unknown type notifications in root handler\n"
    },
    {
      "commit": "9a60ee117bbeaf2fb9a02ea80a6bdbc2811ca4d2",
      "tree": "0f39b8feaec355fd423b909d0ffb0fa0b70d7652",
      "parents": [
        "6dda80ffa4403123f30fa4c77ad948e7907f9bfb",
        "f75d72309192e52ef9a3efb390b1c4f408c142df"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Fri Apr 25 13:12:25 2014 -0700"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Fri Apr 25 13:12:25 2014 -0700"
      },
      "message": "Merge tag \u0027hwmon-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/groeck/linux-staging\n\nPull hwmon fixes from Guenter Roeck:\n - ltc2945: Don\u0027t unecessarily crash kernel on implementation error\n - vexpress: Fix \u0027name\u0027 and \u0027label\u0027 attributes\n\n* tag \u0027hwmon-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/groeck/linux-staging:\n  hwmon: (ltc2945) Don\u0027t crash the kernel unnecessarily\n  hwmon: (vexpress) Avoid creating non-existing attributes\n  hwmon: (vexpress) Use legal hwmon device names\n"
    },
    {
      "commit": "6dda80ffa4403123f30fa4c77ad948e7907f9bfb",
      "tree": "70efeb8da18fb3cfc9047358951959868d1086ae",
      "parents": [
        "3fe89d2e768792a924d3c1e9310ba0b4448cb78e",
        "c69e6f812bab0d5442b40e2f1bfbca48d40bc50b"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Fri Apr 25 13:07:24 2014 -0700"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Fri Apr 25 13:07:24 2014 -0700"
      },
      "message": "Merge tag \u0027scsi-fixes\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi\n\nPull SCSI fixes from James Bottomley:\n \"This is a set of seven fixes, three (hpsa) and free\u0027d command\n  references correcting bugs in the last round of updates and the\n  remaining four correcting problems within the SCSI error handler that\n  was causing a deadlock within USB\"\n\n* tag \u0027scsi-fixes\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi:\n  [SCSI] More USB deadlock fixes\n  [SCSI] Fix USB deadlock caused by SCSI error handling\n  [SCSI] Fix command result state propagation\n  [SCSI] Fix spurious request sense in error handling\n  [SCSI] don\u0027t reference freed command in scsi_prep_return\n  [SCSI] don\u0027t reference freed command in scsi_init_sgtable\n  [SCSI] hpsa: fix NULL dereference in hpsa_put_ctlr_into_performant_mode()\n"
    },
    {
      "commit": "3fe89d2e768792a924d3c1e9310ba0b4448cb78e",
      "tree": "22a8b0ff48f56324516f21afd7965f602b04de92",
      "parents": [
        "625bba662c17e917e164dc37f61aebbc49987ed6",
        "76e7745e8e4330fdb30f049303d524261c0b7a2c"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Fri Apr 25 13:02:02 2014 -0700"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Fri Apr 25 13:02:02 2014 -0700"
      },
      "message": "Merge tag \u0027fixes-3.15-rc3\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/arm/arm-soc\n\nPull ARM SoC fixes from Arnd Bergmann:\n \"Since we didn\u0027t get around to collect fixes in time for -rc2 over the\n  easter vacation, this one is unfortunately a bit larger than we\u0027d like\n  for an -rc3 merge.\n\n  A large set of the changes is in the device tree sources, so I\u0027m\n  splitting out the description between code changes and DT changes.\n  Aside from omap and versatile express, the actual code bugs are and\n  trivial.  Here is an overview:\n\n  imx:\n   - fix video clock settings\n   - fix one clock refcounting bug\n\n  omap:\n   - update defconfig for renamed USB PHY driver\n   - fix error handling in gpmc\n   - fix N900 video initialization regression\n   - fix reression in hwmod code from missing braces\n   - fix am43xx and omap3 clocks\n   - remove bogus write to voltage control register\n\n  pxa:\n   - fix build regression from 3.13 header cleanup\n\n  rockchip:\n   - fix a misleading printk string\n\n  shmobile:\n   - fix incorrect sound setting on multiple machines\n\n  spear:\n   - remove incorrect __init section annotation\n\n  tegra:\n   - remove a stale Kconfig entry\n\n  u300:\n   - update defconfig\n\n  ux500:\n   - enable common wireless and sensor drivers in defconfig\n   - more defconfig updates\n\n  vexpress:\n   - fix voltage calculation for opp\n   - fix reboot hang and warning\n   - fix out-of-bounds array access\n   - improve error handling in clock driver\n\n  overall:\n   - always select CLKSRC_OF in multiplatform builds\n\n  And these are the devicetree related changes:\n\n  imx:\n   - add missing #clock-cell properties\n   - fix pinctrl setting in imx6sl-evk\n   - fix video endpoint on imx53\n   - remove obsolete lvds-channel nodes (multiple patches)\n   - add missing second stmpe node\n   - fix usb host mode on dmo-edmqmx6 (multiple patches)\n   - fix gic node #address-cells to match usage\n   - add missing legacy IRQ map for PCIe\n   - fix microsom pincontrol setting for rgmii\n   - fix fatal typo in touchscreen DT usage for mx5\n   - list all RAM present on m53evk and mx53qsb\n\n  omap:\n   - fix bug in DT handling of gpmc external bus\n   - add DT for older revision of beagleboard\n   - fix regression after DT node name fixes\n   - remove obsolete properties for gpmc\n   - fix pinmux comment to match DT it refers to\n   - fix newly added dra7xx clock node data\n   - add missing clock for USB PHY\n\n  mvebu:\n   - add missing clock for mdio node\n   - fix nonstandard vendor prefixes on i2c nodes\n\n  rockchip:\n   - fix pin control setting for uart\n\n  shmobile:\n   - fix typo in DT data for pin control (multiple patches)\n   - fix gic node #address-cells to match usage\n\n  tegra:\n   - fix clock and uart DT representation to match hardware\n\n  zynq:\n   - add DT nodes for newly added driver\n   - add DT properties required for cpufreq-ondemand\n\n  overall:\n   - restore alphabetic order in Makefile\n   - grammar fixes in bindings\"\n\n* tag \u0027fixes-3.15-rc3\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/arm/arm-soc: (66 commits)\n  ARM: vexpress/TC2: Convert OPP voltage to uV before storing\n  power/reset: vexpress: Fix restart/power off operation\n  dt: tegra: remove non-existent clock IDs\n  clk: tegra: remove non-existent clocks\n  ARM: tegra: remove UART5/UARTE from tegra124.dtsi\n  ARM: tegra: remove TEGRA_EMC_SCALING_ENABLE\n  ARM: Tidy up DTB Makefile entries\n  ARM: fix missing CLKSRC_OF on multi-platform\n  ARM: spear: add __init to spear_clocksource_init()\n  ARM: pxa: hx4700.h: include \"irqs.h\" for PXA_NR_BUILTIN_GPIO\n  arm/mach-vexpress: array accessed out of bounds\n  clk: vexpress: NULL dereference on error path\n  ARM: OMAP2+: Fix GPMC remap for devices using an offset\n  ARM: zynq: dt: Add I2C nodes to Zynq device tree\n  ARM: zynq: DT: Add \u0027clock-latency\u0027 property\n  ARM: OMAP2+: Fix oops for GPMC free\n  ARM: dts: Add support for the BeagleBoard xM A/B\n  ARM: dts: Grammar /that will/it will/\n  ARM: dts: Grammar /is uses/ is used/\n  ARM: OMAP2+: Fix config name for USB3 PHY\n  ...\n"
    },
    {
      "commit": "625bba662c17e917e164dc37f61aebbc49987ed6",
      "tree": "d8a373269a58197e45b932f65b41044390b0071a",
      "parents": [
        "b8e6dece3746902c29a4f0007c062641fa5dcc2d",
        "cff2fce58b2b0f59089e7edcdc38803d65057b9f"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Fri Apr 25 12:40:32 2014 -0700"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Fri Apr 25 12:40:32 2014 -0700"
      },
      "message": "Merge tag \u0027locks-v3.15-2\u0027 of git://git.samba.org/jlayton/linux\n\nPull file locking fixes from Jeff Layton:\n \"File locking related bugfixes for v3.15 (pile #2)\n\n   - fix for a long-standing bug in __break_lease that can cause soft\n     lockups\n   - renaming of file-private locks to \"open file description\" locks,\n     and the command macros to more visually distinct names\n\n  The fix for __break_lease is also in the pile of patches for which\n  Bruce sent a pull request, but I assume that your merge procedure will\n  handle that correctly.\n\n  For the other patches, I don\u0027t like the fact that we need to rename\n  this stuff at this late stage, but it should be settled now\n  (hopefully)\"\n\n* tag \u0027locks-v3.15-2\u0027 of git://git.samba.org/jlayton/linux:\n  locks: rename FL_FILE_PVT and IS_FILE_PVT to use \"*_OFDLCK\" instead\n  locks: rename file-private locks to \"open file description locks\"\n  locks: allow __break_lease to sleep even when break_time is 0\n"
    },
    {
      "commit": "b8e6dece3746902c29a4f0007c062641fa5dcc2d",
      "tree": "dcd5c7a70a04f89448cffd408a1743f90f41e7a6",
      "parents": [
        "76429f1dedbc1815d875c9b68eefd30c95cb8fa9",
        "fc208d026be0c7d60db9118583fc62f6ca97743d"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Fri Apr 25 12:39:05 2014 -0700"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Fri Apr 25 12:39:05 2014 -0700"
      },
      "message": "Merge branch \u0027for-3.15\u0027 of git://linux-nfs.org/~bfields/linux\n\nPull nfsd bugfixes from Bruce Fields:\n \"Three small nfsd bugfixes (including one locks.c fix for a bug\n  triggered only from nfsd).\n\n  Jeff\u0027s patches are for long-existing problems that became easier to\n  trigger since the addition of vfs delegation support\"\n\n* \u0027for-3.15\u0027 of git://linux-nfs.org/~bfields/linux:\n  Revert \"nfsd4: fix nfs4err_resource in 4.1 case\"\n  nfsd: set timeparms.to_maxval in setup_callback_client\n  locks: allow __break_lease to sleep even when break_time is 0\n"
    },
    {
      "commit": "0c8c77d35582c3f7989f1316368da5ae7f14ad4b",
      "tree": "0a166093a460b69fc1ec02e6f216708155f70710",
      "parents": [
        "b44b2140265ddfde03acbe809336111d31adb0d1"
      ],
      "author": {
        "name": "Christian Borntraeger",
        "email": "borntraeger@de.ibm.com",
        "time": "Wed Apr 23 20:58:45 2014 +0200"
      },
      "committer": {
        "name": "Greg Kroah-Hartman",
        "email": "gregkh@linuxfoundation.org",
        "time": "Fri Apr 25 12:25:14 2014 -0700"
      },
      "message": "s390/ccwgroup: Fix memory corruption\n\ncommit 0b60f9ead5d4816e7e3d6e28f4a0d22d4a1b2513 (s390: use\ndevice_remove_file_self() instead of device_schedule_callback())\n\ncaused random memory corruption on my s390 box. Turns out that the\nlast element of the ccwgroup structure is of dynamic size, so we\nmust move the newly introduced work structure _before_ the zero\nlength array.\n\nSigned-off-by: Christian Borntraeger \u003cborntraeger@de.ibm.com\u003e\nCC: Tejun Heo \u003ctj@kernel.org\u003e\nCC: Martin Schwidefsky \u003cschwidefsky@de.ibm.com\u003e\nCC: Heiko Carstens \u003cheiko.carstens@de.ibm.com\u003e\nCC: Sebastian Ott \u003csebott@linux.vnet.ibm.com\u003e\nCC: Peter Oberparleiter \u003coberpar@linux.vnet.ibm.com\u003e\nAcked-by: Tejun Heo \u003ctj@kernel.org\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n"
    },
    {
      "commit": "b44b2140265ddfde03acbe809336111d31adb0d1",
      "tree": "816005d20738096829c9730747eba6373ce3cd04",
      "parents": [
        "c1befb885939cdaaf420c10bbe9ff57aa00446ea"
      ],
      "author": {
        "name": "Tejun Heo",
        "email": "tj@kernel.org",
        "time": "Sun Apr 20 08:29:21 2014 -0400"
      },
      "committer": {
        "name": "Greg Kroah-Hartman",
        "email": "gregkh@linuxfoundation.org",
        "time": "Fri Apr 25 12:25:13 2014 -0700"
      },
      "message": "kernfs: add back missing error check in kernfs_fop_mmap()\n\nWhile updating how mmap enabled kernfs files are handled by lockdep,\n9b2db6e18945 (\"sysfs: bail early from kernfs_file_mmap() to avoid\nspurious lockdep warning\") inadvertently dropped error return check\nfrom kernfs_file_mmap().  The intention was just dropping \"if\n(ops-\u003emmap)\" check as the control won\u0027t reach the point if the mmap\ncallback isn\u0027t implemented, but I mistakenly removed the error return\ncheck together with it.\n\nThis led to Xorg crash on i810 which was reported and bisected to the\ncommit and then to the specific change by Tobias.\n\nSigned-off-by: Tejun Heo \u003ctj@kernel.org\u003e\nReported-and-bisected-by: Tobias Powalowski \u003ctobias.powalowski@googlemail.com\u003e\nTested-by: Tobias Powalowski \u003ctobias.powalowski@googlemail.com\u003e\nReferences: http://lkml.kernel.org/g/533D01BD.1010200@googlemail.com\nCc: stable \u003cstable@vger.kernel.org\u003e # 3.14\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n"
    },
    {
      "commit": "c1befb885939cdaaf420c10bbe9ff57aa00446ea",
      "tree": "e46f5797e3b17838b41f874f70facaf7b52bfea5",
      "parents": [
        "a798c10faf62a505d24e5f6213fbaf904a39623f"
      ],
      "author": {
        "name": "Jianyu Zhan",
        "email": "nasa4836@gmail.com",
        "time": "Thu Apr 17 17:52:10 2014 +0800"
      },
      "committer": {
        "name": "Greg Kroah-Hartman",
        "email": "gregkh@linuxfoundation.org",
        "time": "Fri Apr 25 12:25:13 2014 -0700"
      },
      "message": "kernfs: fix a subdir count leak\n\nCurrently kernfs_link_sibling() increates parent-\u003edir.subdirs before\nadding the node into parent\u0027s chidren rb tree.\n\nBecause it is possible that kernfs_link_sibling() couldn\u0027t find\na suitable slot and bail out, this leads to a mismatch between\nelevated subdir count with actual children node numbers.\n\nThis patches fix this problem, by moving the subdir accouting\nafter the actual addtion happening.\n\nSigned-off-by: Jianyu Zhan \u003cnasa4836@gmail.com\u003e\nAcked-by: Tejun Heo \u003ctj@kernel.org\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n"
    },
    {
      "commit": "34f972d6156fe9eea2ab7bb418c71f9d1d5c8e7b",
      "tree": "31b9587194c12d35c242c518f9d742c71318bde1",
      "parents": [
        "dd6b48ecec2ea7d15f28d5e5474388681899a5e1"
      ],
      "author": {
        "name": "Bjørn Mork",
        "email": "bjorn@mork.no",
        "time": "Fri Apr 25 18:49:20 2014 +0200"
      },
      "committer": {
        "name": "Greg Kroah-Hartman",
        "email": "gregkh@linuxfoundation.org",
        "time": "Fri Apr 25 11:35:04 2014 -0700"
      },
      "message": "usb: option: add and update a number of CMOTech devices\n\nA number of older CMOTech modems are based on Qualcomm\nchips.  The blacklisted interfaces are QMI/wwan.\n\nReported-by: Lars Melin \u003clarsm17@gmail.com\u003e\nCc: \u003cstable@vger.kernel.org\u003e\nSigned-off-by: Bjørn Mork \u003cbjorn@mork.no\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n"
    },
    {
      "commit": "dd6b48ecec2ea7d15f28d5e5474388681899a5e1",
      "tree": "f32324c39a84cc894c1a94baf8eb648e4ac0150f",
      "parents": [
        "533b3994610f316e5cd61b56d0c4daa15c830f89"
      ],
      "author": {
        "name": "Bjørn Mork",
        "email": "bjorn@mork.no",
        "time": "Fri Apr 25 18:49:19 2014 +0200"
      },
      "committer": {
        "name": "Greg Kroah-Hartman",
        "email": "gregkh@linuxfoundation.org",
        "time": "Fri Apr 25 11:35:04 2014 -0700"
      },
      "message": "usb: option: add Alcatel L800MA\n\nDevice interface layout:\n0: ff/ff/ff - serial\n1: ff/00/00 - serial AT+PPP\n2: ff/ff/ff - QMI/wwan\n3: 08/06/50 - storage\n\nCc: \u003cstable@vger.kernel.org\u003e\nSigned-off-by: Bjørn Mork \u003cbjorn@mork.no\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n"
    },
    {
      "commit": "533b3994610f316e5cd61b56d0c4daa15c830f89",
      "tree": "dcb3ffa610a7613828747dddfb093c2c636f7723",
      "parents": [
        "bce4f588f19d59fc07fadfeb0b2a3a06c942827a"
      ],
      "author": {
        "name": "Bjørn Mork",
        "email": "bjorn@mork.no",
        "time": "Fri Apr 25 18:49:18 2014 +0200"
      },
      "committer": {
        "name": "Greg Kroah-Hartman",
        "email": "gregkh@linuxfoundation.org",
        "time": "Fri Apr 25 11:35:04 2014 -0700"
      },
      "message": "usb: option: add Olivetti Olicard 500\n\nDevice interface layout:\n0: ff/ff/ff - serial\n1: ff/ff/ff - serial AT+PPP\n2: 08/06/50 - storage\n3: ff/ff/ff - serial\n4: ff/ff/ff - QMI/wwan\n\nCc: \u003cstable@vger.kernel.org\u003e\nReported-by: Julio Araujo \u003cjulio.araujo@wllctel.com.br\u003e\nSigned-off-by: Bjørn Mork \u003cbjorn@mork.no\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n"
    },
    {
      "commit": "bce4f588f19d59fc07fadfeb0b2a3a06c942827a",
      "tree": "92d808a8f73a25919e28011c5ae7e204a21f5e16",
      "parents": [
        "70a3615fc07c2330ed7c1e922f3c44f4a67c0762"
      ],
      "author": {
        "name": "Bjørn Mork",
        "email": "bjorn@mork.no",
        "time": "Fri Apr 25 18:49:17 2014 +0200"
      },
      "committer": {
        "name": "Greg Kroah-Hartman",
        "email": "gregkh@linuxfoundation.org",
        "time": "Fri Apr 25 11:35:04 2014 -0700"
      },
      "message": "usb: qcserial: add Sierra Wireless MC7305/MC7355\n\nCc: \u003cstable@vger.kernel.org\u003e\nSigned-off-by: Bjørn Mork \u003cbjorn@mork.no\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n"
    },
    {
      "commit": "70a3615fc07c2330ed7c1e922f3c44f4a67c0762",
      "tree": "c68bf303ba2c1e614b04ccd466ef50c3efed2f22",
      "parents": [
        "a00986f81182a69dee4d2c48e8c19805bdf0f790"
      ],
      "author": {
        "name": "Bjørn Mork",
        "email": "bjorn@mork.no",
        "time": "Fri Apr 25 18:49:16 2014 +0200"
      },
      "committer": {
        "name": "Greg Kroah-Hartman",
        "email": "gregkh@linuxfoundation.org",
        "time": "Fri Apr 25 11:35:04 2014 -0700"
      },
      "message": "usb: qcserial: add Sierra Wireless MC73xx\n\nCc: \u003cstable@vger.kernel.org\u003e\nSigned-off-by: Bjørn Mork \u003cbjorn@mork.no\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n"
    },
    {
      "commit": "a00986f81182a69dee4d2c48e8c19805bdf0f790",
      "tree": "a6b92110603a501283ceab8b07c4231e46478a9b",
      "parents": [
        "5509076d1b4485ce9fb07705fcbcd2695907ab5b"
      ],
      "author": {
        "name": "Bjørn Mork",
        "email": "bjorn@mork.no",
        "time": "Fri Apr 25 18:49:15 2014 +0200"
      },
      "committer": {
        "name": "Greg Kroah-Hartman",
        "email": "gregkh@linuxfoundation.org",
        "time": "Fri Apr 25 11:35:04 2014 -0700"
      },
      "message": "usb: qcserial: add Sierra Wireless EM7355\n\nCc: \u003cstable@vger.kernel.org\u003e\nSigned-off-by: Bjørn Mork \u003cbjorn@mork.no\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n"
    },
    {
      "commit": "bc3ee18a7a57243721ecfd879319e3d2e882f289",
      "tree": "ffec556fa31106c8d8e6a21647e8c0449b4fdd25",
      "parents": [
        "8f0712037b4ed63dfce844939ac9866054f15ca0"
      ],
      "author": {
        "name": "Chanho Min",
        "email": "chanho.min@lge.com",
        "time": "Mon Apr 14 08:38:53 2014 +0100"
      },
      "committer": {
        "name": "Catalin Marinas",
        "email": "catalin.marinas@arm.com",
        "time": "Fri Apr 25 18:15:56 2014 +0100"
      },
      "message": "arm64: init: Move of_clk_init to time_init\n\nClock providers should be initialized before clocksource_of_init.\nIf not, Clock source initialization can be fail to get the clock.\n\nAcked-by: Will Deacon \u003cwill.deacon@arm.com\u003e\nSigned-off-by: Chanho Min \u003cchanho.min@lge.com\u003e\nSigned-off-by: Will Deacon \u003cwill.deacon@arm.com\u003e\nSigned-off-by: Catalin Marinas \u003ccatalin.marinas@arm.com\u003e\n"
    },
    {
      "commit": "5509076d1b4485ce9fb07705fcbcd2695907ab5b",
      "tree": "8b60716a6f575040d3cc10d97d97fb891a9b9f18",
      "parents": [
        "01bb59ebffdec314da8da66266edf29529372f9b"
      ],
      "author": {
        "name": "Johan Hovold",
        "email": "jhovold@gmail.com",
        "time": "Fri Apr 25 15:23:03 2014 +0200"
      },
      "committer": {
        "name": "Greg Kroah-Hartman",
        "email": "gregkh@linuxfoundation.org",
        "time": "Fri Apr 25 09:34:10 2014 -0700"
      },
      "message": "USB: io_ti: fix firmware download on big-endian machines\n\nDuring firmware download the device expects memory addresses in\nbig-endian byte order. As the wIndex parameter which hold the address is\nsent in little-endian byte order regardless of host byte order, we need\nto use swab16 rather than cpu_to_be16.\n\nAlso make sure to handle the struct ti_i2c_desc size parameter which is\nreturned in little-endian byte order.\n\nReported-by: Ludovic Drolez \u003cldrolez@debian.org\u003e\nTested-by: Ludovic Drolez \u003cldrolez@debian.org\u003e\nCc: stable@vger.kernel.org\nSigned-off-by: Johan Hovold \u003cjhovold@gmail.com\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n"
    },
    {
      "commit": "01bb59ebffdec314da8da66266edf29529372f9b",
      "tree": "48d094aa52706d2c1af9492562f5e5b6b749edf3",
      "parents": [
        "6db249ebefc6bf5c39f35dfaacc046d8ad3ffd70"
      ],
      "author": {
        "name": "David Cohen",
        "email": "david.a.cohen@linux.intel.com",
        "time": "Fri Apr 25 19:20:16 2014 +0300"
      },
      "committer": {
        "name": "Greg Kroah-Hartman",
        "email": "gregkh@linuxfoundation.org",
        "time": "Fri Apr 25 09:34:10 2014 -0700"
      },
      "message": "usb/xhci: fix compilation warning when !CONFIG_PCI \u0026\u0026 !CONFIG_PM\n\nWhen CONFIG_PCI and CONFIG_PM are not selected, xhci.c gets this\nwarning:\ndrivers/usb/host/xhci.c:409:13: warning: ‘xhci_msix_sync_irqs’ defined\nbut not used [-Wunused-function]\n\nInstead of creating nested #ifdefs, this patch fixes it by defining the\nxHCI PCI stubs as inline.\n\nThis warning has been in since 3.2 kernel and was\ncaused by commit 421aa841a134f6a743111cf44d0c6d3b45e3cf8c\n\"usb/xhci: hide MSI code behind PCI bars\", but wasn\u0027t noticed\nuntil 3.13 when a configuration with these options was tried\n\nSigned-off-by: David Cohen \u003cdavid.a.cohen@linux.intel.com\u003e\nCc: stable@vger.kernel.org # 3.2\nSigned-off-by: Mathias Nyman \u003cmathias.nyman@linux.intel.com\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n"
    },
    {
      "commit": "6db249ebefc6bf5c39f35dfaacc046d8ad3ffd70",
      "tree": "6e6f32d478f62b4c5e0449b9999cdc58774a90c4",
      "parents": [
        "c09ec25d3684cad74d851c0f028a495999591279"
      ],
      "author": {
        "name": "Igor Gnatenko",
        "email": "i.gnatenko.brain@gmail.com",
        "time": "Fri Apr 25 19:20:15 2014 +0300"
      },
      "committer": {
        "name": "Greg Kroah-Hartman",
        "email": "gregkh@linuxfoundation.org",
        "time": "Fri Apr 25 09:34:10 2014 -0700"
      },
      "message": "xhci: extend quirk for Renesas cards\n\nAfter suspend another Renesas PCI-X USB 3.0 card doesn\u0027t work.\n[root@fedora-20 ~]# lspci -vmnnd 1912:\nDevice:\t03:00.0\nClass:\tUSB controller [0c03]\nVendor:\tRenesas Technology Corp. [1912]\nDevice:\tuPD720202 USB 3.0 Host Controller [0015]\nSVendor:\tRenesas Technology Corp. [1912]\nSDevice:\tuPD720202 USB 3.0 Host Controller [0015]\nRev:\t02\nProgIf:\t30\n\nThis patch should be applied to stable kernel 3.14 that contain\nthe commit 1aa9578c1a9450fb21501c4f549f5b1edb557e6d\n\"xhci: Fix resume issues on Renesas chips in Samsung laptops\"\n\nReported-and-tested-by: Anatoly Kharchenko \u003crfr-bugs@yandex.ru\u003e\nReference: http://redmine.russianfedora.pro/issues/1315\nSigned-off-by: Igor Gnatenko \u003ci.gnatenko.brain@gmail.com\u003e\nCc: stable@vger.kernel.org # 3.14\nSigned-off-by: Mathias Nyman \u003cmathias.nyman@linux.intel.com\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n"
    },
    {
      "commit": "c09ec25d3684cad74d851c0f028a495999591279",
      "tree": "f99b2d634dc438bd1775ba40f73ea0229457f4e0",
      "parents": [
        "1f81b6d22a5980955b01e08cf27fb745dc9b686f"
      ],
      "author": {
        "name": "Denis Turischev",
        "email": "denis.turischev@compulab.co.il",
        "time": "Fri Apr 25 19:20:14 2014 +0300"
      },
      "committer": {
        "name": "Greg Kroah-Hartman",
        "email": "gregkh@linuxfoundation.org",
        "time": "Fri Apr 25 09:34:10 2014 -0700"
      },
      "message": "xhci: Switch Intel Lynx Point ports to EHCI on shutdown.\n\nThe same issue like with Panther Point chipsets. If the USB ports are\nswitched to xHCI on shutdown, the xHCI host will send a spurious interrupt,\nwhich will wake the system. Some BIOS have work around for this, but not all.\nOne example is Compulab\u0027s mini-desktop, the Intense-PC2.\n\nThe bug can be avoided if the USB ports are switched back to EHCI on\nshutdown.\n\nThis patch should be backported to stable kernels as old as 3.12,\nthat contain the commit 638298dc66ea36623dbc2757a24fc2c4ab41b016\n\"xhci: Fix spurious wakeups after S5 on Haswell\"\n\nSigned-off-by: Denis Turischev \u003cdenis@compulab.co.il\u003e\nCc: stable@vger.kernel.org\nSigned-off-by: Mathias Nyman \u003cmathias.nyman@linux.intel.com\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n"
    },
    {
      "commit": "1f81b6d22a5980955b01e08cf27fb745dc9b686f",
      "tree": "c0159cc1d616db9e3aed055e076c3c7197ca9b8e",
      "parents": [
        "2b97789fa289d531e767d994a77e34ec58f328c4"
      ],
      "author": {
        "name": "Julius Werner",
        "email": "jwerner@chromium.org",
        "time": "Fri Apr 25 19:20:13 2014 +0300"
      },
      "committer": {
        "name": "Greg Kroah-Hartman",
        "email": "gregkh@linuxfoundation.org",
        "time": "Fri Apr 25 09:34:10 2014 -0700"
      },
      "message": "usb: xhci: Prefer endpoint context dequeue pointer over stopped_trb\n\nWe have observed a rare cycle state desync bug after Set TR Dequeue\nPointer commands on Intel LynxPoint xHCs (resulting in an endpoint that\ndoesn\u0027t fetch new TRBs and thus an unresponsive USB device). It always\ntriggers when a previous Set TR Dequeue Pointer command has set the\npointer to the final Link TRB of a segment, and then another URB gets\nenqueued and cancelled again before it can be completed. Further\ninvestigation showed that the xHC had returned the Link TRB in the TRB\nPointer field of the Transfer Event (CC \u003d\u003d Stopped -- Length Invalid),\nbut when xhci_find_new_dequeue_state() later accesses the Endpoint\nContext\u0027s TR Dequeue Pointer field it is set to the first TRB of the\nnext segment.\n\nThe driver expects those two values to be the same in this situation,\nand uses the cycle state of the latter together with the address of the\nformer. This should be fine according to the XHCI specification, since\nthe endpoint ring should be stopped when returning the Transfer Event\nand thus should not advance over the Link TRB before it gets restarted.\nHowever, real-world XHCI implementations apparently don\u0027t really care\nthat much about these details, so the driver should follow a more\ndefensive approach to try to work around HC spec violations.\n\nThis patch removes the stopped_trb variable that had been used to store\nthe TRB Pointer from the last Transfer Event of a stopped TRB. Instead,\nxhci_find_new_dequeue_state() now relies only on the Endpoint Context,\nrequiring a small amount of additional processing to find the virtual\naddress corresponding to the TR Dequeue Pointer. Some other parts of the\nfunction were slightly rearranged to better fit into this model.\n\nThis patch should be backported to kernels as old as 2.6.31 that contain\nthe commit ae636747146ea97efa18e04576acd3416e2514f5 \"USB: xhci: URB\ncancellation support.\"\n\nSigned-off-by: Julius Werner \u003cjwerner@chromium.org\u003e\nCc: stable@vger.kernel.org\nSigned-off-by: Mathias Nyman \u003cmathias.nyman@linux.intel.com\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n"
    },
    {
      "commit": "8f0712037b4ed63dfce844939ac9866054f15ca0",
      "tree": "52d5337b100611342949226e6be2d9369c5e6a61",
      "parents": [
        "43683afbcb32f7b7318ac1badd6469d91fe22711"
      ],
      "author": {
        "name": "Leo Yan",
        "email": "leoy@marvell.com",
        "time": "Wed Apr 16 13:26:35 2014 +0100"
      },
      "committer": {
        "name": "Catalin Marinas",
        "email": "catalin.marinas@arm.com",
        "time": "Fri Apr 25 16:46:51 2014 +0100"
      },
      "message": "arm64: initialize spinlock for init_mm\u0027s context\n\nARM64 has defined the spinlock for init_mm\u0027s context, so need initialize\nthe spinlock structure; otherwise during the suspend flow it will dump\nthe info for spinlock\u0027s bad magic warning as below:\n\n[   39.084394] Disabling non-boot CPUs ...\n[   39.092871] BUG: spinlock bad magic on CPU#1, swapper/1/0\n[   39.092896]  lock: init_mm+0x338/0x3e0, .magic: 00000000, .owner: \u003cnone\u003e/-1, .owner_cpu: 0\n[   39.092907] CPU: 1 PID: 0 Comm: swapper/1 Tainted: G           O 3.10.33 #125\n[   39.092912] Call trace:\n[   39.092927] [\u003cffffffc000087e64\u003e] dump_backtrace+0x0/0x16c\n[   39.092934] [\u003cffffffc000087fe0\u003e] show_stack+0x10/0x1c\n[   39.092947] [\u003cffffffc000765334\u003e] dump_stack+0x1c/0x28\n[   39.092953] [\u003cffffffc0007653b8\u003e] spin_dump+0x78/0x88\n[   39.092960] [\u003cffffffc0007653ec\u003e] spin_bug+0x24/0x34\n[   39.092971] [\u003cffffffc000300a28\u003e] do_raw_spin_lock+0x98/0x17c\n[   39.092979] [\u003cffffffc00076cf08\u003e] _raw_spin_lock_irqsave+0x4c/0x60\n[   39.092990] [\u003cffffffc000094044\u003e] set_mm_context+0x1c/0x6c\n[   39.092996] [\u003cffffffc0000941c8\u003e] __new_context+0x94/0x10c\n[   39.093007] [\u003cffffffc0000d63d4\u003e] idle_task_exit+0x104/0x1b0\n[   39.093014] [\u003cffffffc00008d91c\u003e] cpu_die+0x14/0x74\n[   39.093021] [\u003cffffffc000084f74\u003e] arch_cpu_idle_dead+0x8/0x14\n[   39.093030] [\u003cffffffc0000e7f18\u003e] cpu_startup_entry+0x1ec/0x258\n[   39.093036] [\u003cffffffc00008d810\u003e] secondary_start_kernel+0x114/0x124\n\nSigned-off-by: Leo Yan \u003cleoy@marvell.com\u003e\nAcked-by: Will Deacon \u003cwill.deacon@arm.com\u003e\nSigned-off-by: Catalin Marinas \u003ccatalin.marinas@arm.com\u003e\n"
    },
    {
      "commit": "43683afbcb32f7b7318ac1badd6469d91fe22711",
      "tree": "7278aa89b4da8148393092d5944b45284a0a2590",
      "parents": [
        "2eb835e058c737205d35d9a8791ad27b0f9e89a4"
      ],
      "author": {
        "name": "Will Deacon",
        "email": "will.deacon@arm.com",
        "time": "Thu Apr 17 12:37:14 2014 +0100"
      },
      "committer": {
        "name": "Catalin Marinas",
        "email": "catalin.marinas@arm.com",
        "time": "Fri Apr 25 16:46:50 2014 +0100"
      },
      "message": "arm64: debug: remove noisy, pointless warning\n\nSending a SIGTRAP to a user task after execution of a BRK instruction at\nEL0 is fundamental to the way in which software breakpoints work and\ndoesn\u0027t deserve a warning to be logged in dmesg. Whilst the warning can\nbe justified from EL1, do_debug_exception will already do the right thing,\nso simply remove the code altogether.\n\nCc: Sandeepa Prabhu \u003csandeepa.prabhu@linaro.org\u003e\nReported-by: Kyrylo Tkachov \u003ckyrylo.tkachov@arm.com\u003e\nSigned-off-by: Will Deacon \u003cwill.deacon@arm.com\u003e\nSigned-off-by: Catalin Marinas \u003ccatalin.marinas@arm.com\u003e\n"
    },
    {
      "commit": "2eb835e058c737205d35d9a8791ad27b0f9e89a4",
      "tree": "ee87c206a0f1c55db2ff6fdfcfd5f044d47cb2e4",
      "parents": [
        "c46a98ab1045565db6d858b628fbfde80f05281a"
      ],
      "author": {
        "name": "Steve Capper",
        "email": "steve.capper@linaro.org",
        "time": "Thu Apr 24 15:33:21 2014 +0100"
      },
      "committer": {
        "name": "Catalin Marinas",
        "email": "catalin.marinas@arm.com",
        "time": "Fri Apr 25 16:46:49 2014 +0100"
      },
      "message": "arm64: mm: Add THP TLB entries to general mmu_gather\n\nWhen arm64 moved over to the core mmu_gather, it lost the logic to\nflush THP TLB entries (tlb_remove_pmd_tlb_entry was removed and the\ncore implementation only signals that the mmu_gather needs a flush).\n\nThis patch ensures that tlb_add_flush is called for THP TLB entries.\n\nSigned-off-by: Steve Capper \u003csteve.capper@linaro.org\u003e\nAcked-by: Will Deacon \u003cwill.deacon@arm.com\u003e\nSigned-off-by: Catalin Marinas \u003ccatalin.marinas@arm.com\u003e\n"
    },
    {
      "commit": "d93003e8e4e1fbbc8a06ec561a63f5aa105a4c45",
      "tree": "d6ef909b8363db3597e8c28b7483f94a7eea7c25",
      "parents": [
        "cd1711709fe98d3569e7f8215ba31adcfc3686ae"
      ],
      "author": {
        "name": "Sebastian Hesselbarth",
        "email": "sebastian.hesselbarth@gmail.com",
        "time": "Thu Apr 24 22:58:30 2014 +0100"
      },
      "committer": {
        "name": "Russell King",
        "email": "rmk+kernel@arm.linux.org.uk",
        "time": "Fri Apr 25 12:07:35 2014 +0100"
      },
      "message": "ARM: 8042/1: iwmmxt: allow to build iWMMXt on Marvell PJ4B\n\nSome Marvell PJ4B CPUs also implement iWMMXt extensions. With a\nproper check for iWMMXt coprocessors now in place, enable it by\ndefault on PJ4B. While at it, also allow to manually select\nthe corresponding Kconfig option.\n\nSigned-off-by: Sebastian Hesselbarth \u003csebastian.hesselbarth@gmail.com\u003e\nTested-by: Thomas Petazzoni \u003cthomas.petazzoni@free-electrons.com\u003e\nTested-by: Kevin Hilman \u003ckhilman@linaro.org\u003e\nSigned-off-by: Russell King \u003crmk+kernel@arm.linux.org.uk\u003e\n"
    },
    {
      "commit": "cd1711709fe98d3569e7f8215ba31adcfc3686ae",
      "tree": "34c2e7c7314c3dbe9db3c29227b54ae6589a115d",
      "parents": [
        "e89f443b182c7a813a60d85dbf22d090231b5e6b"
      ],
      "author": {
        "name": "Sebastian Hesselbarth",
        "email": "sebastian.hesselbarth@gmail.com",
        "time": "Thu Apr 24 22:58:00 2014 +0100"
      },
      "committer": {
        "name": "Russell King",
        "email": "rmk+kernel@arm.linux.org.uk",
        "time": "Fri Apr 25 12:07:34 2014 +0100"
      },
      "message": "ARM: 8041/1: pj4: fix cpu_is_pj4 check\n\nCommit fdb487f5c961b94486a78fa61fa28b8eff1954ab\n  (\"ARM: 8015/1: Add cpu_is_pj4 to distinguish PJ4 because it\n    has some differences with V7\")\nintroduced a cpuid check for Marvell PJ4 processors to fix a\nregression caused by adding PJ4 based Marvell Dove into\nmulti_v7.\n\nUnfortunately, this check is too narrow to catch PJ4 used on\nDove itself and breaks iWMMXt support.\n\nThis patch therefore relaxes the cpuid mask to match both PJ4\nand PJ4B. Also, rework the given comment about PJ4/PJ4B\nmodifications to be a little bit more specific about the\ndifferences.\n\nSigned-off-by: Sebastian Hesselbarth \u003csebastian.hesselbarth@gmail.com\u003e\nTested-by: Thomas Petazzoni \u003cthomas.petazzoni@free-electrons.com\u003e\nTested-by: Kevin Hilman \u003ckhilman@linaro.org\u003e\nSigned-off-by: Russell King \u003crmk+kernel@arm.linux.org.uk\u003e\n"
    },
    {
      "commit": "e89f443b182c7a813a60d85dbf22d090231b5e6b",
      "tree": "c43e0fcc04db4e7cab75de84f7ddf10c217fadb8",
      "parents": [
        "7d0656598924b2781fb96cdff566c9f844643fcd"
      ],
      "author": {
        "name": "Sebastian Hesselbarth",
        "email": "sebastian.hesselbarth@gmail.com",
        "time": "Thu Apr 24 22:57:25 2014 +0100"
      },
      "committer": {
        "name": "Russell King",
        "email": "rmk+kernel@arm.linux.org.uk",
        "time": "Fri Apr 25 12:07:34 2014 +0100"
      },
      "message": "ARM: 8040/1: pj4: properly detect existence of iWMMXt coprocessor\n\ncommit fdb487f5c961b94486a78fa61fa28b8eff1954ab\n  (\"ARM: 8015/1: Add cpu_is_pj4 to distinguish PJ4 because it\n    has some differences with V7\")\nintroduced a fix for checking PJ4 cpuid to not use PJ4 specific\ncoprocessor access on non-PJ4 platforms.\n\nUnfortunately, this in turn broke Marvell Armada 370/XP, both\ncomprising Marvell PJ4B CPUs without iWMMXt extension. Instead\nof only checking for cpuid, which may not be sufficient to\ndetermine iWMMXt support, the presence of iWMMXt coprocessors\ncan be checked by enabling and reading the Coprocessor ID\nregister (wCID, register 0 of CP1).\n\nTherefore this adds an explicit check for the presence and correct\nwCID value, before enabling iWMMXt capabilities. As a bonus, also\nprint the iWMMXt version of a detected coprocessor.\n\nThis has been tested to properly detect iWMMXt presence/absence on:\n- PJ4,  CPUID 0x560f5815, wCID 0x56052001: Marvell Dove, iWMMXt v2\n- PJ4B, CPUID 0x561f5811: Marvell Armada 370, no iWMMXt\n- PJ4B, CPUID 0x562f5841, wCID 0x56052001: Marvell Armada 1500, iWMMXt v2\n- PJ4B, CPUID 0x562f5842: Marvell Armada XP, no iWMMXt\n\nSigned-off-by: Sebastian Hesselbarth \u003csebastian.hesselbarth@gmail.com\u003e\nTested-by: Thomas Petazzoni \u003cthomas.petazzoni@free-electrons.com\u003e\nTested-by: Kevin Hilman \u003ckhilman@linaro.org\u003e\nSigned-off-by: Russell King \u003crmk+kernel@arm.linux.org.uk\u003e\n"
    },
    {
      "commit": "7d0656598924b2781fb96cdff566c9f844643fcd",
      "tree": "a47f983783b2cce7d72666715e7509edf3262c79",
      "parents": [
        "c2f07fe64dc0da6d4ccc02d858695356bd685aeb"
      ],
      "author": {
        "name": "Sebastian Hesselbarth",
        "email": "sebastian.hesselbarth@gmail.com",
        "time": "Thu Apr 24 22:56:43 2014 +0100"
      },
      "committer": {
        "name": "Russell King",
        "email": "rmk+kernel@arm.linux.org.uk",
        "time": "Fri Apr 25 12:07:33 2014 +0100"
      },
      "message": "ARM: 8039/1: pj4: enable iWMMXt only if CONFIG_IWMMXT is set\n\nThis fixes PJ4 coprocessor init to only expose iWMMXt capabilities,\nif the corresponding kernel support for iWMMXt is enabled.\n\nSigned-off-by: Sebastian Hesselbarth \u003csebastian.hesselbarth@gmail.com\u003e\nTested-by: Thomas Petazzoni \u003cthomas.petazzoni@free-electrons.com\u003e\nTested-by: Kevin Hilman \u003ckhilman@linaro.org\u003e\nSigned-off-by: Russell King \u003crmk+kernel@arm.linux.org.uk\u003e\n"
    },
    {
      "commit": "c2f07fe64dc0da6d4ccc02d858695356bd685aeb",
      "tree": "90016fd73c4038c0b2de98e92412fa6833b63174",
      "parents": [
        "556d3f7f4d791cae54fd24ef28296e666f4c96a6"
      ],
      "author": {
        "name": "Sebastian Hesselbarth",
        "email": "sebastian.hesselbarth@gmail.com",
        "time": "Thu Apr 24 22:54:58 2014 +0100"
      },
      "committer": {
        "name": "Russell King",
        "email": "rmk+kernel@arm.linux.org.uk",
        "time": "Fri Apr 25 12:07:32 2014 +0100"
      },
      "message": "ARM: 8038/1: iwmmxt: explicitly check for supported architectures\n\niwmmxt.S requires special treatment of coprocessor access registers\nfor PJ4 and XScale-based CPUs. It only checks for CPU_PJ4 and drops\ndown to XScale-based treatment on all other architectures.\n\nAs some PJ4B also come with iWMMXt and also need PJ4 treatment,\nrework the corresponding preprocessor directives to explicitly\ncheck for supported architectures and fail on unsupported ones.\n\nSigned-off-by: Sebastian Hesselbarth \u003csebastian.hesselbarth@gmail.com\u003e\nTested-by: Thomas Petazzoni \u003cthomas.petazzoni@free-electrons.com\u003e\nTested-by: Kevin Hilman \u003ckhilman@linaro.org\u003e\nSigned-off-by: Russell King \u003crmk+kernel@arm.linux.org.uk\u003e\n"
    },
    {
      "commit": "76e7745e8e4330fdb30f049303d524261c0b7a2c",
      "tree": "648153b454922b722030944a8c845152499ae955",
      "parents": [
        "1fc52762e33cc905331681364d79424d921f60f2",
        "0f6faa3fc909482c2b40161de9bcf0d5460e54c5"
      ],
      "author": {
        "name": "Arnd Bergmann",
        "email": "arnd@arndb.de",
        "time": "Fri Apr 25 11:22:20 2014 +0200"
      },
      "committer": {
        "name": "Arnd Bergmann",
        "email": "arnd@arndb.de",
        "time": "Fri Apr 25 11:22:20 2014 +0200"
      },
      "message": "Merge tag \u0027zynq-dt-fixes-for-3.15\u0027 of git://git.xilinx.com/linux-xlnx into fixes\n\narm: Xilinx Zynq DT fixes for v3.15\n\n- Enable Zynq I2c\n- Fix cpufreq DT binding\n\n* tag \u0027zynq-dt-fixes-for-3.15\u0027 of git://git.xilinx.com/linux-xlnx:\n  ARM: zynq: dt: Add I2C nodes to Zynq device tree\n  ARM: zynq: DT: Add \u0027clock-latency\u0027 property\n\nSigned-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\n"
    },
    {
      "commit": "f8213bdc89719bad895a02c62c4a85066ff76720",
      "tree": "bae57bba7903618c8831d860474fb6052bb90310",
      "parents": [
        "9ce49a0b4ff7f13961d8d106ffae959823d2e758"
      ],
      "author": {
        "name": "Filipe Manana",
        "email": "fdmanana@gmail.com",
        "time": "Thu Apr 24 15:15:29 2014 +0100"
      },
      "committer": {
        "name": "Chris Mason",
        "email": "clm@fb.com",
        "time": "Thu Apr 24 16:43:33 2014 -0700"
      },
      "message": "Btrfs: correctly set profile flags on seqlock retry\n\nIf we had to retry on the profiles seqlock (due to a concurrent write), we\nwould set bits on the input flags that corresponded both to the current\nprofile and to previous values of the profile.\n\nSigned-off-by: Filipe David Borba Manana \u003cfdmanana@gmail.com\u003e\nSigned-off-by: Chris Mason \u003cclm@fb.com\u003e\n"
    },
    {
      "commit": "9ce49a0b4ff7f13961d8d106ffae959823d2e758",
      "tree": "b7abb6948560dbe6f21af61bdd62d90be7b3735e",
      "parents": [
        "1c70d8fb4dfa95bee491816b2a6767b5ca1080e7"
      ],
      "author": {
        "name": "Filipe Manana",
        "email": "fdmanana@gmail.com",
        "time": "Thu Apr 24 15:15:28 2014 +0100"
      },
      "committer": {
        "name": "Chris Mason",
        "email": "clm@fb.com",
        "time": "Thu Apr 24 16:43:33 2014 -0700"
      },
      "message": "Btrfs: use correct key when repeating search for extent item\n\nIf skinny metadata is enabled and our first tree search fails to find a\nskinny extent item, we may repeat a tree search for a \"fat\" extent item\n(if the previous item in the leaf is not the \"fat\" extent we\u0027re looking\nfor). However we were not setting the new key\u0027s objectid to the right\nvalue, as we previously used the same key variable to peek at the previous\nitem in the leaf, which has a different objectid. So just set the right\nobjectid to avoid modifying/deleting a wrong item if we repeat the tree\nsearch.\n\nSigned-off-by: Filipe David Borba Manana \u003cfdmanana@gmail.com\u003e\nSigned-off-by: Chris Mason \u003cclm@fb.com\u003e\n"
    },
    {
      "commit": "1c70d8fb4dfa95bee491816b2a6767b5ca1080e7",
      "tree": "c866788e0d25fb2f2d1621b1e2e651319a8114b9",
      "parents": [
        "28c16cbbc32781224309e50cc99c684f2498bc59"
      ],
      "author": {
        "name": "Miao Xie",
        "email": "miaox@cn.fujitsu.com",
        "time": "Wed Apr 23 19:33:36 2014 +0800"
      },
      "committer": {
        "name": "Chris Mason",
        "email": "clm@fb.com",
        "time": "Thu Apr 24 16:43:33 2014 -0700"
      },
      "message": "Btrfs: fix inode caching vs tree log\n\nCurrently, with inode cache enabled, we will reuse its inode id immediately\nafter unlinking file, we may hit something like following:\n\n|-\u003eiput inode\n|-\u003ereturn inode id into inode cache\n|-\u003ecreate dir,fsync\n|-\u003epower off\n\nAn easy way to reproduce this problem is:\n\nmkfs.btrfs -f /dev/sdb\nmount /dev/sdb /mnt -o inode_cache,commit\u003d100\ndd if\u003d/dev/zero of\u003d/mnt/data bs\u003d1M count\u003d10 oflag\u003dsync\ninode_id\u003d`ls -i /mnt/data | awk \u0027{print $1}\u0027`\nrm -f /mnt/data\n\ni\u003d1\nwhile [ 1 ]\ndo\n        mkdir /mnt/dir_$i\n        test1\u003d`stat /mnt/dir_$i | grep Inode: | awk \u0027{print $4}\u0027`\n        if [ $test1 -eq $inode_id ]\n        then\n\t\tdd if\u003d/dev/zero of\u003d/mnt/dir_$i/data bs\u003d1M count\u003d1 oflag\u003dsync\n\t\techo b \u003e /proc/sysrq-trigger\n\tfi\n\tsleep 1\n        i\u003d$(($i+1))\ndone\n\nmount /dev/sdb /mnt\numount /dev/sdb\nbtrfs check /dev/sdb\n\nWe fix this problem by adding unlinked inode\u0027s id into pinned tree,\nand we can not reuse them until committing transaction.\n\nCc: stable@vger.kernel.org\nSigned-off-by: Miao Xie \u003cmiaox@cn.fujitsu.com\u003e\nSigned-off-by: Wang Shilong \u003cwangsl.fnst@cn.fujitsu.com\u003e\nSigned-off-by: Chris Mason \u003cclm@fb.com\u003e\n"
    },
    {
      "commit": "28c16cbbc32781224309e50cc99c684f2498bc59",
      "tree": "5dab03fd6e797956a396bb231fc7e2e136cd69b9",
      "parents": [
        "e60efa84252c059bde5f65fccc6af94478d39e3b"
      ],
      "author": {
        "name": "Wang Shilong",
        "email": "wangsl.fnst@cn.fujitsu.com",
        "time": "Wed Apr 23 19:33:35 2014 +0800"
      },
      "committer": {
        "name": "Chris Mason",
        "email": "clm@fb.com",
        "time": "Thu Apr 24 16:43:32 2014 -0700"
      },
      "message": "Btrfs: fix possible memory leaks in open_ctree()\n\nFix possible memory leaks in the following error handling paths:\n\nread_tree_block()\nbtrfs_recover_log_trees\nbtrfs_commit_super()\nbtrfs_find_orphan_roots()\nbtrfs_cleanup_fs_roots()\n\nSigned-off-by: Wang Shilong \u003cwangsl.fnst@cn.fujitsu.com\u003e\nSigned-off-by: Chris Mason \u003cclm@fb.com\u003e\n"
    },
    {
      "commit": "e60efa84252c059bde5f65fccc6af94478d39e3b",
      "tree": "5dc4a09dfda089e777d67a29c3255236931ede60",
      "parents": [
        "9d89ce658718d9d21465666127a15ca2c82c4ea7"
      ],
      "author": {
        "name": "Wang Shilong",
        "email": "wangsl.fnst@cn.fujitsu.com",
        "time": "Wed Apr 23 19:33:34 2014 +0800"
      },
      "committer": {
        "name": "Chris Mason",
        "email": "clm@fb.com",
        "time": "Thu Apr 24 16:43:32 2014 -0700"
      },
      "message": "Btrfs: avoid triggering bug_on() when we fail to start inode caching task\n\nWhen running stress test(including snapshots,balance,fstress), we trigger\nthe following BUG_ON() which is because we fail to start inode caching task.\n\n[  181.131945] kernel BUG at fs/btrfs/inode-map.c:179!\n[  181.137963] invalid opcode: 0000 [#1] SMP\n[  181.217096] CPU: 11 PID: 2532 Comm: btrfs Not tainted 3.14.0 #1\n[  181.240521] task: ffff88013b621b30 ti: ffff8800b6ada000 task.ti: ffff8800b6ada000\n[  181.367506] Call Trace:\n[  181.371107]  [\u003cffffffffa036c1be\u003e] btrfs_return_ino+0x9e/0x110 [btrfs]\n[  181.379191]  [\u003cffffffffa038082b\u003e] btrfs_evict_inode+0x46b/0x4c0 [btrfs]\n[  181.387464]  [\u003cffffffff810b5a70\u003e] ? autoremove_wake_function+0x40/0x40\n[  181.395642]  [\u003cffffffff811dc5fe\u003e] evict+0x9e/0x190\n[  181.401882]  [\u003cffffffff811dcde3\u003e] iput+0xf3/0x180\n[  181.408025]  [\u003cffffffffa03812de\u003e] btrfs_orphan_cleanup+0x1ee/0x430 [btrfs]\n[  181.416614]  [\u003cffffffffa03a6abd\u003e] btrfs_mksubvol.isra.29+0x3bd/0x450 [btrfs]\n[  181.425399]  [\u003cffffffffa03a6cd6\u003e] btrfs_ioctl_snap_create_transid+0x186/0x190 [btrfs]\n[  181.435059]  [\u003cffffffffa03a6e3b\u003e] btrfs_ioctl_snap_create_v2+0xeb/0x130 [btrfs]\n[  181.444148]  [\u003cffffffffa03a9656\u003e] btrfs_ioctl+0xf76/0x2b90 [btrfs]\n[  181.451971]  [\u003cffffffff8117e565\u003e] ? handle_mm_fault+0x475/0xe80\n[  181.459509]  [\u003cffffffff8167ba0c\u003e] ? __do_page_fault+0x1ec/0x520\n[  181.467046]  [\u003cffffffff81185b35\u003e] ? do_mmap_pgoff+0x2f5/0x3c0\n[  181.474393]  [\u003cffffffff811d4da8\u003e] do_vfs_ioctl+0x2d8/0x4b0\n[  181.481450]  [\u003cffffffff811d5001\u003e] SyS_ioctl+0x81/0xa0\n[  181.488021]  [\u003cffffffff81680b69\u003e] system_call_fastpath+0x16/0x1b\n\nWe should avoid triggering BUG_ON() here, instead, we output warning messages\nand clear inode_cache option.\n\nSigned-off-by: Wang Shilong \u003cwangsl.fnst@cn.fujitsu.com\u003e\nSigned-off-by: Chris Mason \u003cclm@fb.com\u003e\n"
    },
    {
      "commit": "9d89ce658718d9d21465666127a15ca2c82c4ea7",
      "tree": "31a53b58e98f8c63a9f1e7bba0c673b35cf1d743",
      "parents": [
        "3f9e3df8da3c51649c15db249978a10f7374236a"
      ],
      "author": {
        "name": "Wang Shilong",
        "email": "wangsl.fnst@cn.fujitsu.com",
        "time": "Wed Apr 23 19:33:33 2014 +0800"
      },
      "committer": {
        "name": "Chris Mason",
        "email": "clm@fb.com",
        "time": "Thu Apr 24 16:43:32 2014 -0700"
      },
      "message": "Btrfs: move btrfs_{set,clear}_and_info() to ctree.h\n\nSigned-off-by: Wang Shilong \u003cwangsl.fnst@cn.fujitsu.com\u003e\nSigned-off-by: Chris Mason \u003cclm@fb.com\u003e\n"
    },
    {
      "commit": "3f9e3df8da3c51649c15db249978a10f7374236a",
      "tree": "27c95d85e319fe75e4643fdeaf79ae928bbef6b3",
      "parents": [
        "c5f7d0bb29df2e1848a236e58e201daf5b4e0f21"
      ],
      "author": {
        "name": "David Sterba",
        "email": "dsterba@suse.cz",
        "time": "Tue Apr 15 18:50:17 2014 +0200"
      },
      "committer": {
        "name": "Chris Mason",
        "email": "clm@fb.com",
        "time": "Thu Apr 24 16:43:32 2014 -0700"
      },
      "message": "btrfs: replace error code from btrfs_drop_extents\n\nThere\u0027s a case which clone does not handle and used to BUG_ON instead,\n(testcase xfstests/btrfs/035), now returns EINVAL. This error code is\nconfusing to the ioctl caller, as it normally signifies errorneous\narguments.\n\nChange it to ENOPNOTSUPP which allows a fall back to copy instead of\nclone. This does not affect the common reflink operation.\n\nSigned-off-by: David Sterba \u003cdsterba@suse.cz\u003e\nSigned-off-by: Chris Mason \u003cclm@fb.com\u003e\n"
    },
    {
      "commit": "c5f7d0bb29df2e1848a236e58e201daf5b4e0f21",
      "tree": "00190de63e0573193302449e3df0d13cc467ac41",
      "parents": [
        "0040e606e35a0db80fc3fac04ccc7c7176a8e2b1"
      ],
      "author": {
        "name": "Qu Wenruo",
        "email": "quwenruo@cn.fujitsu.com",
        "time": "Tue Apr 15 10:41:00 2014 +0800"
      },
      "committer": {
        "name": "Chris Mason",
        "email": "clm@fb.com",
        "time": "Thu Apr 24 16:43:32 2014 -0700"
      },
      "message": "btrfs: Change the hole range to a more accurate value.\n\nCommit 3ac0d7b96a268a98bd474cab8bce3a9f125aaccf fixed the btrfs expanding\nwrite problem but the hole punched is sometimes too large for some\niovec, which has unmapped data ranges.\nThis patch will change to hole range to a more accurate value using the\ncounts checked by the write check routines.\n\nReported-by: Al Viro \u003cviro@ZenIV.linux.org.uk\u003e\nSigned-off-by: Qu Wenruo \u003cquwenruo@cn.fujitsu.com\u003e\nSigned-off-by: Chris Mason \u003cclm@fb.com\u003e\n"
    },
    {
      "commit": "7deb39ed8d9494ea541bcaa69b56036a94f79dc2",
      "tree": "8f7f242c45c98cd06200846e23142b363594cb5a",
      "parents": [
        "f94b0572683716f727b25086f8d39671506593ab"
      ],
      "author": {
        "name": "Thomas Pfaff",
        "email": "tpfaff@pcs.com",
        "time": "Wed Apr 23 12:33:22 2014 +0200"
      },
      "committer": {
        "name": "Greg Kroah-Hartman",
        "email": "gregkh@linuxfoundation.org",
        "time": "Thu Apr 24 15:38:22 2014 -0700"
      },
      "message": "serial_core: fix uart PORT_UNKNOWN handling\n\nWhile porting a RS485 driver from 2.6.29 to 3.14, i noticed that the serial tty\ndriver could break it by using uart ports that it does not own :\n\n1. uart_change_pm ist called during uart_open and calls the uart pm function\n   without checking for PORT_UNKNOWN.\n   The fix is to move uart_change_pm from uart_open to uart_port_startup.\n2. The return code from the uart request_port call in uart_set_info is not\n   handled properly, leading to the situation that the serial driver also\n   thinks it owns the uart ports.\n   This can triggered by doing following actions :\n\n   setserial /dev/ttyS0 uart none    # release the uart ports\n   modprobe lirc-serial              # or any other device that uses the uart\n   setserial /dev/ttyS0 uart 16550   # gives no error and the uart tty driver\n                                     # can use the ports as well\n\nSigned-off-by: Thomas Pfaff \u003ctpfaff@pcs.com\u003e\nReviewed-by: Alan Cox \u003calan@linux.intel.com\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n"
    },
    {
      "commit": "f94b0572683716f727b25086f8d39671506593ab",
      "tree": "c1428aa4245d01149c008c70558b15a166c84d12",
      "parents": [
        "ab88c8dc3bd667bcf21c8319165db92b930cf7e7"
      ],
      "author": {
        "name": "Doug Anderson",
        "email": "dianders@chromium.org",
        "time": "Mon Apr 21 09:40:36 2014 -0700"
      },
      "committer": {
        "name": "Greg Kroah-Hartman",
        "email": "gregkh@linuxfoundation.org",
        "time": "Thu Apr 24 15:38:22 2014 -0700"
      },
      "message": "serial: samsung: Change barrier() to cpu_relax() in console output\n\nThe two functions to write out to the console (one used in normal\nconsole mode and one in polling console mode) were slightly different.\nOne used a barrier() in its loop and the other a cpu_relax().  The\nbarrier() really doesn\u0027t do anything since we\u0027re using rd_regl() to\nread the port anyway.  Switch it to cpu_relax() to make things\nconsistent.\n\nNo known bugs / issues are fixed by this change--it just makes things\nmore consistent.\n\nSigned-off-by: Doug Anderson \u003cdianders@chromium.org\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n"
    },
    {
      "commit": "ab88c8dc3bd667bcf21c8319165db92b930cf7e7",
      "tree": "060c678b26d846eab33e1861862c0191e27e269f",
      "parents": [
        "bb7f09ba961dd43a2398975cc2d4ad0eb77ec865"
      ],
      "author": {
        "name": "Doug Anderson",
        "email": "dianders@chromium.org",
        "time": "Mon Apr 21 09:40:35 2014 -0700"
      },
      "committer": {
        "name": "Greg Kroah-Hartman",
        "email": "gregkh@linuxfoundation.org",
        "time": "Thu Apr 24 15:38:21 2014 -0700"
      },
      "message": "serial: samsung: don\u0027t check config for every character\n\nThe s3c24xx_serial_console_putchar() is _only_ ever used by\ns3c24xx_serial_console_write() and is called in a loop (indirectly\nthrough uart_console_write()).  There\u0027s no reason to call\ns3c24xx_port_configured() for every iteration through the loop.  Move\nit outside the loop.\n\nSigned-off-by: Doug Anderson \u003cdianders@chromium.org\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n"
    },
    {
      "commit": "bb7f09ba961dd43a2398975cc2d4ad0eb77ec865",
      "tree": "9e5f6702835084d1dcad66929272fddc08a3e620",
      "parents": [
        "f8fd1b0350d3a4581125f5eda6528f5a2c5f9183"
      ],
      "author": {
        "name": "Doug Anderson",
        "email": "dianders@chromium.org",
        "time": "Mon Apr 21 09:40:34 2014 -0700"
      },
      "committer": {
        "name": "Greg Kroah-Hartman",
        "email": "gregkh@linuxfoundation.org",
        "time": "Thu Apr 24 15:38:21 2014 -0700"
      },
      "message": "serial: samsung: Use the passed in \"port\", fixing kgdb w/ no console\n\nThe two functions in the samsung serial driver used for writing\ncharacters out to the port were inconsistent about whether they used\nthe passed in \"port\" or the global \"cons_uart\".  There was no reason\nto use the global and the use of the global in\ns3c24xx_serial_put_poll_char() caused a crash in the case where you\nused the serial port for kgdboc but not for console.\n\nFix it so we used the passed in variable.\n\nNote that this doesn\u0027t fix all problems with the samsung serial\ndriver.  Specifically:\n* s3c24xx_serial_console_putchar() is still 99% identical to\n  s3c24xx_serial_put_poll_char() (the function signature is different,\n  but that\u0027s about it).  A future patch will make them slightly less\n  identical and judging by other serial drivers we may need yet more\n  differences eventually.\n* The samsung serial driver still doesn\u0027t allow you to have more than\n  one console port since it still uses the global cons_uart in\n  s3c24xx_serial_console_write().\n\nSigned-off-by: Doug Anderson \u003cdianders@chromium.org\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n"
    },
    {
      "commit": "f8fd1b0350d3a4581125f5eda6528f5a2c5f9183",
      "tree": "f350e27d766290b1a14ea59d5bc5d40e87f77eb2",
      "parents": [
        "b08c9c317e3f7764a91d522cd031639ba42b98cc"
      ],
      "author": {
        "name": "Loic Poulain",
        "email": "loic.poulain@intel.com",
        "time": "Thu Apr 24 11:34:48 2014 +0200"
      },
      "committer": {
        "name": "Greg Kroah-Hartman",
        "email": "gregkh@linuxfoundation.org",
        "time": "Thu Apr 24 15:18:02 2014 -0700"
      },
      "message": "serial: 8250: Fix thread unsafe __dma_tx_complete function\n\n__dma_tx_complete is not protected against concurrent\ncall of serial8250_tx_dma. it can lead to circular tail\nindex corruption or parallel call of serial_tx_dma on the\nsame data portion.\n\nThis patch fixes this issue by holding the port lock.\n\nSigned-off-by: Loic Poulain \u003cloic.poulain@intel.com\u003e\nReviewed-by: Heikki Krogerus \u003cheikki.krogerus@linux.intel.com\u003e\nCc: stable \u003cstable@vger.kernel.org\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n"
    },
    {
      "commit": "b08c9c317e3f7764a91d522cd031639ba42b98cc",
      "tree": "7d0f1387ed4c14229923b1c8b8239032dcc00173",
      "parents": [
        "6a20dbd6caa2358716136144bf524331d70b1e03"
      ],
      "author": {
        "name": "Loic Poulain",
        "email": "loic.poulain@intel.com",
        "time": "Thu Apr 24 11:38:56 2014 +0200"
      },
      "committer": {
        "name": "Greg Kroah-Hartman",
        "email": "gregkh@linuxfoundation.org",
        "time": "Thu Apr 24 15:18:02 2014 -0700"
      },
      "message": "8250_core: Fix unwanted TX chars write\n\nOn transmit-hold-register empty, serial8250_tx_chars\nshould be called only if we don\u0027t use DMA.\nDMA has its own tx cycle.\n\nSigned-off-by: Loic Poulain \u003cloic.poulain@intel.com\u003e\nReviewed-by: Heikki Krogerus \u003cheikki.krogerus@linux.intel.com\u003e\nCc: stable \u003cstable@vger.kernel.org\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n"
    },
    {
      "commit": "6a20dbd6caa2358716136144bf524331d70b1e03",
      "tree": "c42e7dbd262e07db1ae3680739d5a5c1f0204f09",
      "parents": [
        "a798c10faf62a505d24e5f6213fbaf904a39623f"
      ],
      "author": {
        "name": "Manfred Schlaegl",
        "email": "manfred.schlaegl@gmx.at",
        "time": "Tue Apr 08 14:42:04 2014 +0200"
      },
      "committer": {
        "name": "Greg Kroah-Hartman",
        "email": "gregkh@linuxfoundation.org",
        "time": "Thu Apr 24 15:18:02 2014 -0700"
      },
      "message": "tty: Fix race condition between __tty_buffer_request_room and flush_to_ldisc\n\nThe race was introduced while development of linux-3.11 by\ne8437d7ecbc50198705331449367d401ebb3181f and\ne9975fdec0138f1b2a85b9624e41660abd9865d4.\nOriginally it was found and reproduced on linux-3.12.15 and\nlinux-3.12.15-rt25, by sending 500 byte blocks with 115kbaud to the\ntarget uart in a loop with 100 milliseconds delay.\n\nIn short:\n 1. The consumer flush_to_ldisc is on to remove the head tty_buffer.\n 2. The producer adds a number of bytes, so that a new tty_buffer must\n\tbe allocated and added by __tty_buffer_request_room.\n 3. The consumer removes the head tty_buffer element, without handling\n\tnewly committed data.\n\nDetailed example:\n * Initial buffer:\n   * Head, Tail -\u003e 0: used\u003d250; commit\u003d250; read\u003d240; next\u003dNULL\n * Consumer: \u0027\u0027flush_to_ldisc\u0027\u0027\n   * consumed 10 Byte\n   * buffer:\n     * Head, Tail -\u003e 0: used\u003d250; commit\u003d250; read\u003d250; next\u003dNULL\n{{{\n\t\tcount \u003d head-\u003ecommit - head-\u003eread;\t// count \u003d 0\n\t\tif (!count) {\t\t\t\t// enter\n\t\t\t// INTERRUPTED BY PRODUCER -\u003e\n\t\t\tif (head-\u003enext \u003d\u003d NULL)\n\t\t\t\tbreak;\n\t\t\tbuf-\u003ehead \u003d head-\u003enext;\n\t\t\ttty_buffer_free(port, head);\n\t\t\tcontinue;\n\t\t}\n}}}\n * Producer: tty_insert_flip_... 10 bytes + tty_flip_buffer_push\n   * buffer:\n     * Head, Tail -\u003e 0: used\u003d250; commit\u003d250; read\u003d250; next\u003dNULL\n   * added 6 bytes: head-element filled to maximum.\n     * buffer:\n       * Head, Tail -\u003e 0: used\u003d256; commit\u003d250; read\u003d250; next\u003dNULL\n   * added 4 bytes: __tty_buffer_request_room is called\n     * buffer:\n       * Head -\u003e 0: used\u003d256; commit\u003d256; read\u003d250; next\u003d1\n       * Tail -\u003e 1: used\u003d4; commit\u003d0; read\u003d250 next\u003dNULL\n   * push (tty_flip_buffer_push)\n     * buffer:\n       * Head -\u003e 0: used\u003d256; commit\u003d256; read\u003d250; next\u003d1\n       * Tail -\u003e 1: used\u003d4; commit\u003d4; read\u003d250 next\u003dNULL\n * Consumer\n{{{\n\t\tcount \u003d head-\u003ecommit - head-\u003eread;\n\t\tif (!count) {\n\t\t\t// INTERRUPTED BY PRODUCER \u003c-\n\t\t\tif (head-\u003enext \u003d\u003d NULL)\t\t// -\u003e no break\n\t\t\t\tbreak;\n\t\t\tbuf-\u003ehead \u003d head-\u003enext;\n\t\t\ttty_buffer_free(port, head);\n\t\t\t// ERROR: tty_buffer head freed -\u003e 6 bytes lost\n\t\t\tcontinue;\n\t\t}\n}}}\n\nThis patch reintroduces a spin_lock to protect this case. Perhaps later\na lock-less solution could be found.\n\nSigned-off-by: Manfred Schlaegl \u003cmanfred.schlaegl@gmx.at\u003e\nCc: stable \u003cstable@vger.kernel.org\u003e # 3.11\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n"
    },
    {
      "commit": "1fc52762e33cc905331681364d79424d921f60f2",
      "tree": "d7347407cbbdb7a0565e3b4d09aaf40f0705491a",
      "parents": [
        "9ef1af9ea28c23d0eaed97f7f5142788b6cf570a",
        "cf2e0a73ca9ad376825c013ebaa145608abc27d7"
      ],
      "author": {
        "name": "Arnd Bergmann",
        "email": "arnd@arndb.de",
        "time": "Thu Apr 24 23:46:58 2014 +0200"
      },
      "committer": {
        "name": "Arnd Bergmann",
        "email": "arnd@arndb.de",
        "time": "Thu Apr 24 23:46:58 2014 +0200"
      },
      "message": "Merge tag \u0027vexpress/fixes-for-3.15\u0027 of git://git.linaro.org/people/pawel.moll/linux into fixes\n\nARM Versatile Express fixes for 3.15\n\nThis series contains straight-forward fixes for different\nVersatile Express infrastructure drivers:\n\n- NULL pointer dereference on the error path in the clk driver\n- out of boundary array access in the dcscb driver\n- broken restart/power off implementation\n- mis-interpreted voltage unit in the spc driver\n\n* tag \u0027vexpress/fixes-for-3.15\u0027 of git://git.linaro.org/people/pawel.moll/linux:\n  ARM: vexpress/TC2: Convert OPP voltage to uV before storing\n  power/reset: vexpress: Fix restart/power off operation\n  arm/mach-vexpress: array accessed out of bounds\n  clk: vexpress: NULL dereference on error path\n\nIncludes an update to 3.15-rc2\n\nSigned-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\n"
    },
    {
      "commit": "2b97789fa289d531e767d994a77e34ec58f328c4",
      "tree": "86fed1906885de4609b7b0c7cbbcf45c1cf1816d",
      "parents": [
        "743bb387a1edbf1ebbba6cf77c1af3e488886c39"
      ],
      "author": {
        "name": "Grygorii Strashko",
        "email": "grygorii.strashko@ti.com",
        "time": "Sat Apr 19 08:51:44 2014 +0530"
      },
      "committer": {
        "name": "Greg Kroah-Hartman",
        "email": "gregkh@linuxfoundation.org",
        "time": "Thu Apr 24 12:53:38 2014 -0700"
      },
      "message": "phy: core: make NULL a valid phy reference if !CONFIG_GENERIC_PHY\n\nThis fixes a regression on Keystone 2 platforms caused by patch\n57303488cd37da58263e842de134dc65f7c626d5\n\"usb: dwc3: adapt dwc3 core to use Generic PHY Framework\" which adds\noptional support of generic phy in DWC3 core.\n\nOn Keystone 2 platforms the USB is not working now because\nCONFIG_GENERIC_PHY isn\u0027t set and, as result, Generic PHY APIs stubs\nreturn -ENOSYS always. The log shows:\n dwc3 2690000.dwc3: failed to initialize core\n dwc3: probe of 2690000.dwc3 failed with error -38\n\nHence, fix it by making NULL a valid phy reference in Generic PHY\nAPIs stubs in the same way as it was done by the patch\n04c2facad8fee66c981a51852806d8923336f362 \"drivers: phy: Make NULL\na valid phy reference\".\n\nAcked-by: Felipe Balbi \u003cbalbi@ti.com\u003e\nAcked-by: Santosh Shilimkar \u003csantosh.shilimkar@ti.com\u003e\nSigned-off-by: Grygorii Strashko \u003cgrygorii.strashko@ti.com\u003e\nSigned-off-by: Kishon Vijay Abraham I \u003ckishon@ti.com\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n"
    },
    {
      "commit": "743bb387a1edbf1ebbba6cf77c1af3e488886c39",
      "tree": "af51f66f28ac16b87bd8d4dbe344cfeec4d7905a",
      "parents": [
        "907aa3aa8dbb5437696776d40caeea245c8ba3bd"
      ],
      "author": {
        "name": "Sergei Shtylyov",
        "email": "sergei.shtylyov@cogentembedded.com",
        "time": "Sat Apr 19 08:51:43 2014 +0530"
      },
      "committer": {
        "name": "Greg Kroah-Hartman",
        "email": "gregkh@linuxfoundation.org",
        "time": "Thu Apr 24 12:53:38 2014 -0700"
      },
      "message": "phy: fix kernel oops in phy_lookup()\n\nThe kernel oopses in phy_lookup() due to \u0027phy-\u003einit_data\u0027 being NULL if we\nregister PHYs from a device tree probing driver and then call phy_get() on a\ndevice that has no representation in the device tree (e.g. a PCI device).\nChecking the pointer before dereferening it and skipping an interation if\nit\u0027s NULL prevents this kernel oops.\n\nSigned-off-by: Sergei Shtylyov \u003csergei.shtylyov@cogentembedded.com\u003e\nSigned-off-by: Kishon Vijay Abraham I \u003ckishon@ti.com\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n"
    },
    {
      "commit": "907aa3aa8dbb5437696776d40caeea245c8ba3bd",
      "tree": "dd0960e0018b94c81892164f89658f3062891dcc",
      "parents": [
        "d1481832f1dbb9d10fab27269631a130fa082f03"
      ],
      "author": {
        "name": "Jean Delvare",
        "email": "jdelvare@suse.de",
        "time": "Sat Apr 19 08:51:42 2014 +0530"
      },
      "committer": {
        "name": "Greg Kroah-Hartman",
        "email": "gregkh@linuxfoundation.org",
        "time": "Thu Apr 24 12:53:38 2014 -0700"
      },
      "message": "phy: restore OMAP_CONTROL_PHY dependencies\n\nWhen OMAP_CONTROL_USB was renamed to OMAP_CONTROL_PHY (commit\n14da699b), its dependencies were lost in the process. Nothing in the\ncommit message indicates that this removal was intentional, so I think\nit was by accident and the dependencies should be restored.\n\nSigned-off-by: Jean Delvare \u003cjdelvare@suse.de\u003e\nAcked-by: Roger Quadros \u003crogerq@ti.com\u003e\nCc: Felipe Balbi \u003cbalbi@ti.com\u003e\nSigned-off-by: Kishon Vijay Abraham I \u003ckishon@ti.com\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n"
    },
    {
      "commit": "d1481832f1dbb9d10fab27269631a130fa082f03",
      "tree": "4cec3e3c4624302c25dd6974c811c6f1d9064353",
      "parents": [
        "10164c2ad6d2c16809f6c09e278f946e47801b3a"
      ],
      "author": {
        "name": "Arnd Bergmann",
        "email": "arnd@arndb.de",
        "time": "Sat Apr 19 08:51:41 2014 +0530"
      },
      "committer": {
        "name": "Greg Kroah-Hartman",
        "email": "gregkh@linuxfoundation.org",
        "time": "Thu Apr 24 12:53:38 2014 -0700"
      },
      "message": "phy: exynos: fix building as a module\n\nThe top-level phy-samsung-usb2 driver may be configured as a\nloadable module, which currently causes link errors because\nof the dependency on the exynos{5250,4x12,4210}_usb2_phy_config\nsymbol. Solving this could be achieved by exporting these\nsymbols, but as the SoC-specific parts of the driver are not\ncurrently built as modules, it seems better to just link\neverything into one module and avoid the need for the export.\n\nSigned-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\nAcked-by: Kamil Debski \u003ck.debski@samsung.com\u003e\nSigned-off-by: Kishon Vijay Abraham I \u003ckishon@ti.com\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n"
    },
    {
      "commit": "10164c2ad6d2c16809f6c09e278f946e47801b3a",
      "tree": "4526f21efd4778fd21b588ea70746a8430dc0013",
      "parents": [
        "bd130adacaf8cea179f9a700fb694f5be3b05bf0"
      ],
      "author": {
        "name": "Johan Hovold",
        "email": "jhovold@gmail.com",
        "time": "Wed Apr 23 11:32:19 2014 +0200"
      },
      "committer": {
        "name": "Greg Kroah-Hartman",
        "email": "gregkh@linuxfoundation.org",
        "time": "Thu Apr 24 12:50:56 2014 -0700"
      },
      "message": "USB: serial: fix sysfs-attribute removal deadlock\n\nFix driver new_id sysfs-attribute removal deadlock by making sure to\nnot hold any locks that the attribute operations grab when removing the\nattribute.\n\nSpecifically, usb_serial_deregister holds the table mutex when\nderegistering the driver, which includes removing the new_id attribute.\nThis can lead to a deadlock as writing to new_id increments the\nattribute\u0027s active count before trying to grab the same mutex in\nusb_serial_probe.\n\nThe deadlock can easily be triggered by inserting a sleep in\nusb_serial_deregister and writing the id of an unbound device to new_id\nduring module unload.\n\nAs the table mutex (in this case) is used to prevent subdriver unload\nduring probe, it should be sufficient to only hold the lock while\nmanipulating the usb-serial driver list during deregister. A racing\nprobe will then either fail to find a matching subdriver or fail to get\nthe corresponding module reference.\n\nSince v3.15-rc1 this also triggers the following lockdep warning:\n\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\n[ INFO: possible circular locking dependency detected ]\n3.15.0-rc2 #123 Tainted: G        W\n-------------------------------------------------------\nmodprobe/190 is trying to acquire lock:\n (s_active#4){++++.+}, at: [\u003cc0167aa0\u003e] kernfs_remove_by_name_ns+0x4c/0x94\n\nbut task is already holding lock:\n (table_lock){+.+.+.}, at: [\u003cbf004d84\u003e] usb_serial_deregister+0x3c/0x78 [usbserial]\n\nwhich lock already depends on the new lock.\n\nthe existing dependency chain (in reverse order) is:\n\n-\u003e #1 (table_lock){+.+.+.}:\n       [\u003cc0075f84\u003e] __lock_acquire+0x1694/0x1ce4\n       [\u003cc0076de8\u003e] lock_acquire+0xb4/0x154\n       [\u003cc03af3cc\u003e] _raw_spin_lock+0x4c/0x5c\n       [\u003cc02bbc24\u003e] usb_store_new_id+0x14c/0x1ac\n       [\u003cbf007eb4\u003e] new_id_store+0x68/0x70 [usbserial]\n       [\u003cc025f568\u003e] drv_attr_store+0x30/0x3c\n       [\u003cc01690e0\u003e] sysfs_kf_write+0x5c/0x60\n       [\u003cc01682c0\u003e] kernfs_fop_write+0xd4/0x194\n       [\u003cc010881c\u003e] vfs_write+0xbc/0x198\n       [\u003cc0108e4c\u003e] SyS_write+0x4c/0xa0\n       [\u003cc000f880\u003e] ret_fast_syscall+0x0/0x48\n\n-\u003e #0 (s_active#4){++++.+}:\n       [\u003cc03a7a28\u003e] print_circular_bug+0x68/0x2f8\n       [\u003cc0076218\u003e] __lock_acquire+0x1928/0x1ce4\n       [\u003cc0076de8\u003e] lock_acquire+0xb4/0x154\n       [\u003cc0166b70\u003e] __kernfs_remove+0x254/0x310\n       [\u003cc0167aa0\u003e] kernfs_remove_by_name_ns+0x4c/0x94\n       [\u003cc0169fb8\u003e] remove_files.isra.1+0x48/0x84\n       [\u003cc016a2fc\u003e] sysfs_remove_group+0x58/0xac\n       [\u003cc016a414\u003e] sysfs_remove_groups+0x34/0x44\n       [\u003cc02623b8\u003e] driver_remove_groups+0x1c/0x20\n       [\u003cc0260e9c\u003e] bus_remove_driver+0x3c/0xe4\n       [\u003cc026235c\u003e] driver_unregister+0x38/0x58\n       [\u003cbf007fb4\u003e] usb_serial_bus_deregister+0x84/0x88 [usbserial]\n       [\u003cbf004db4\u003e] usb_serial_deregister+0x6c/0x78 [usbserial]\n       [\u003cbf005330\u003e] usb_serial_deregister_drivers+0x2c/0x4c [usbserial]\n       [\u003cbf016618\u003e] usb_serial_module_exit+0x14/0x1c [sierra]\n       [\u003cc009d6cc\u003e] SyS_delete_module+0x184/0x210\n       [\u003cc000f880\u003e] ret_fast_syscall+0x0/0x48\n\nother info that might help us debug this:\n\n Possible unsafe locking scenario:\n\n       CPU0                    CPU1\n       ----                    ----\n  lock(table_lock);\n                               lock(s_active#4);\n                               lock(table_lock);\n  lock(s_active#4);\n\n *** DEADLOCK ***\n\n1 lock held by modprobe/190:\n #0:  (table_lock){+.+.+.}, at: [\u003cbf004d84\u003e] usb_serial_deregister+0x3c/0x78 [usbserial]\n\nstack backtrace:\nCPU: 0 PID: 190 Comm: modprobe Tainted: G        W     3.15.0-rc2 #123\n[\u003cc0015e10\u003e] (unwind_backtrace) from [\u003cc0013728\u003e] (show_stack+0x20/0x24)\n[\u003cc0013728\u003e] (show_stack) from [\u003cc03a9a54\u003e] (dump_stack+0x24/0x28)\n[\u003cc03a9a54\u003e] (dump_stack) from [\u003cc03a7cac\u003e] (print_circular_bug+0x2ec/0x2f8)\n[\u003cc03a7cac\u003e] (print_circular_bug) from [\u003cc0076218\u003e] (__lock_acquire+0x1928/0x1ce4)\n[\u003cc0076218\u003e] (__lock_acquire) from [\u003cc0076de8\u003e] (lock_acquire+0xb4/0x154)\n[\u003cc0076de8\u003e] (lock_acquire) from [\u003cc0166b70\u003e] (__kernfs_remove+0x254/0x310)\n[\u003cc0166b70\u003e] (__kernfs_remove) from [\u003cc0167aa0\u003e] (kernfs_remove_by_name_ns+0x4c/0x94)\n[\u003cc0167aa0\u003e] (kernfs_remove_by_name_ns) from [\u003cc0169fb8\u003e] (remove_files.isra.1+0x48/0x84)\n[\u003cc0169fb8\u003e] (remove_files.isra.1) from [\u003cc016a2fc\u003e] (sysfs_remove_group+0x58/0xac)\n[\u003cc016a2fc\u003e] (sysfs_remove_group) from [\u003cc016a414\u003e] (sysfs_remove_groups+0x34/0x44)\n[\u003cc016a414\u003e] (sysfs_remove_groups) from [\u003cc02623b8\u003e] (driver_remove_groups+0x1c/0x20)\n[\u003cc02623b8\u003e] (driver_remove_groups) from [\u003cc0260e9c\u003e] (bus_remove_driver+0x3c/0xe4)\n[\u003cc0260e9c\u003e] (bus_remove_driver) from [\u003cc026235c\u003e] (driver_unregister+0x38/0x58)\n[\u003cc026235c\u003e] (driver_unregister) from [\u003cbf007fb4\u003e] (usb_serial_bus_deregister+0x84/0x88 [usbserial])\n[\u003cbf007fb4\u003e] (usb_serial_bus_deregister [usbserial]) from [\u003cbf004db4\u003e] (usb_serial_deregister+0x6c/0x78 [usbserial])\n[\u003cbf004db4\u003e] (usb_serial_deregister [usbserial]) from [\u003cbf005330\u003e] (usb_serial_deregister_drivers+0x2c/0x4c [usbserial])\n[\u003cbf005330\u003e] (usb_serial_deregister_drivers [usbserial]) from [\u003cbf016618\u003e] (usb_serial_module_exit+0x14/0x1c [sierra])\n[\u003cbf016618\u003e] (usb_serial_module_exit [sierra]) from [\u003cc009d6cc\u003e] (SyS_delete_module+0x184/0x210)\n[\u003cc009d6cc\u003e] (SyS_delete_module) from [\u003cc000f880\u003e] (ret_fast_syscall+0x0/0x48)\n\nSigned-off-by: Johan Hovold \u003cjhovold@gmail.com\u003e\nCc: stable \u003cstable@vger.kernel.org\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n"
    },
    {
      "commit": "bd130adacaf8cea179f9a700fb694f5be3b05bf0",
      "tree": "16afa8c2512d525c3d5acb371f4194c907a2ae8b",
      "parents": [
        "7584f2ebc18b6355f21be5fb2f75afbf3f781ce5"
      ],
      "author": {
        "name": "Thomas Pugliese",
        "email": "thomas.pugliese@gmail.com",
        "time": "Wed Apr 23 14:28:10 2014 -0500"
      },
      "committer": {
        "name": "Greg Kroah-Hartman",
        "email": "gregkh@linuxfoundation.org",
        "time": "Thu Apr 24 12:45:41 2014 -0700"
      },
      "message": "usb: wusbcore: fix panic in wusbhc_chid_set\n\nIf no valid CHID value has previously been set on an HWA, writing a\nvalue of all zeros will cause a kernel panic in uwb_radio_stop because\nwusbhc-\u003euwb_rc has not been set.  This patch skips the call to\nuwb_radio_stop if wusbhc-\u003euwb_rc has not been initialized.\n\nSigned-off-by: Thomas Pugliese \u003cthomas.pugliese@gmail.com\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n"
    },
    {
      "commit": "7584f2ebc18b6355f21be5fb2f75afbf3f781ce5",
      "tree": "048f31e5c303f4fcbb1cd5d7110598ffeec72d0f",
      "parents": [
        "c996b9379180ee578b9ea208bb4f197201bbb023"
      ],
      "author": {
        "name": "Thomas Pugliese",
        "email": "thomas.pugliese@gmail.com",
        "time": "Wed Apr 23 14:32:27 2014 -0500"
      },
      "committer": {
        "name": "Greg Kroah-Hartman",
        "email": "gregkh@linuxfoundation.org",
        "time": "Thu Apr 24 12:45:40 2014 -0700"
      },
      "message": "usb: wusbcore: convert nested lock to use spin_lock instead of spin_lock_irq\n\nNesting a spin_lock_irq/unlock_irq inside a lock that has already\ndisabled interrupts will enable interrupts before we are ready when\nspin_unlock_irq is called.  This patch converts the inner lock to use\nspin_lock and spin_unlock instead.\n\nSigned-off-by: Thomas Pugliese \u003cthomas.pugliese@gmail.com\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n"
    },
    {
      "commit": "c996b9379180ee578b9ea208bb4f197201bbb023",
      "tree": "f0dc1591a2906562c00505c08c392360aa8c6db7",
      "parents": [
        "cd84f009e98b3aa6109503f80f66ccf3e19e8fa9"
      ],
      "author": {
        "name": "Thomas Pugliese",
        "email": "thomas.pugliese@gmail.com",
        "time": "Wed Apr 23 14:42:47 2014 -0500"
      },
      "committer": {
        "name": "Greg Kroah-Hartman",
        "email": "gregkh@linuxfoundation.org",
        "time": "Thu Apr 24 12:45:40 2014 -0700"
      },
      "message": "uwb: don\u0027t call spin_unlock_irq in a USB completion handler\n\nThis patch converts the use of spin_lock_irq/spin_unlock_irq to\nspin_lock_irqsave/spin_unlock_irqrestore in uwb_rc_set_drp_cmd_done\nwhich is called from a USB completion handler.  There are also\nwhitespace cleanups to make checkpatch.pl happy.\n\nSigned-off-by: Thomas Pugliese \u003cthomas.pugliese@gmail.com\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n"
    },
    {
      "commit": "cd84f009e98b3aa6109503f80f66ccf3e19e8fa9",
      "tree": "cea706c4b4d11c0e225716d49474aabafd234710",
      "parents": [
        "e988f306e7241a0adddb650357d34b6537d77b0e"
      ],
      "author": {
        "name": "Peter Chen",
        "email": "peter.chen@freescale.com",
        "time": "Thu Apr 24 08:15:27 2014 +0800"
      },
      "committer": {
        "name": "Greg Kroah-Hartman",
        "email": "gregkh@linuxfoundation.org",
        "time": "Thu Apr 24 12:45:40 2014 -0700"
      },
      "message": "usb: chipidea: coordinate usb phy initialization for different phy type\n\nFor internal PHY (like UTMI), the phy clock may from internal pll,\nit is on/off on the fly, the access PORTSC.PTS will hang without\nphy clock. So, the usb_phy_init which will open phy clock needs to\nbe called before hw_phymode_configure.\nSee: http://marc.info/?l\u003dlinux-arm-kernel\u0026m\u003d139350618732108\u0026w\u003d2\n\nFor external PHY (like ulpi), it needs to configure portsc.pts before\nvisit viewport, or the viewport can\u0027t be visited. so phy_phymode_configure\nneeds to be called before usb_phy_init.\nSee: cd0b42c2a6d2a74244f0053f8960f5dad5842278\n\nIt may not the best solution, but it can work for all situations.\n\nCc: Sascha Hauer \u003cs.hauer@pengutronix.de\u003e\nCc: Chris Ruehl \u003cchris.ruehl@gtsys.com.hk\u003e\nCc: shc_work@mail.ru\nCc: denis@eukrea.com\nCc: festevam@gmail.com\nCc: stable \u003cstable@vger.kernel.org\u003e # 3.14\nSigned-off-by: Peter Chen \u003cpeter.chen@freescale.com\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n"
    },
    {
      "commit": "e988f306e7241a0adddb650357d34b6537d77b0e",
      "tree": "6f13d31af2ec7518785ce73dcb6b850f70964f17",
      "parents": [
        "a798c10faf62a505d24e5f6213fbaf904a39623f",
        "a31a942a148e0083ce560ffeb54fb60e06ab7201"
      ],
      "author": {
        "name": "Greg Kroah-Hartman",
        "email": "gregkh@linuxfoundation.org",
        "time": "Thu Apr 24 12:33:10 2014 -0700"
      },
      "committer": {
        "name": "Greg Kroah-Hartman",
        "email": "gregkh@linuxfoundation.org",
        "time": "Thu Apr 24 12:33:10 2014 -0700"
      },
      "message": "Merge tag \u0027fixes-for-v3.15-rc3\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/balbi/usb into usb-linus\n\nFelipe writes:\n\nusb: fixes for v3.15-rc3\n\nQuite a few fixes this time since I lost v3.15-rc2\nwindow.\n\nMost fixes are MUSB which learned to remove its debugfs directories\nproperly, got a fix for PHY handling and now knows that it should\nmake sure its clocks aren\u0027t gated before trying to access registers.\n\nffs got a race fix between ffs_epfile_io() and ffs_func_eps_disable().\n\ndwc3 got a fix for system suspend/resume and now only iterates over\nvalid endpoints when trying to resize TX fifos.\n\nusb_get_phy() now will properly return an error if try_module_get() fails.\n\nWe also have a revert for a NAPI conversion on the ethernet gadget which\nwas causing a kernel BUG.\n\nSigned-of-by: Felipe Balbi \u003cbalbi@ti.com\u003e\n"
    },
    {
      "commit": "76429f1dedbc1815d875c9b68eefd30c95cb8fa9",
      "tree": "0c7efa089ad10e6c503b9c3a5ccc6735ceb11ad3",
      "parents": [
        "ff1e5b447e465da75dfdd0a4f4070cfef86a4155",
        "38962571375149626cd282ef9d8e87e8faf21155"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Thu Apr 24 12:01:58 2014 -0700"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Thu Apr 24 12:01:58 2014 -0700"
      },
      "message": "Merge tag \u0027regulator-v3.15-rc2\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/broonie/regulator\n\nPull regulator fixes from Mark Brown:\n \"A couple of things here:\n\n   - Fixes for pbias that didn\u0027t make it in during the merge window due\n     to the driver coming in via MMC.  The conversion to use helpers is\n     a fix as it implements list_voltage() which the main user (MMC)\n     relies on for correct functioning.\n   - Change the !REGULATOR stub for optional regulators to return an\n     error rather than a dummy; this is more in keeping with the\n     intended use of optional regulators and fixes some issues seen MMC\n     where it got confused by a dummy being provided\"\n\n* tag \u0027regulator-v3.15-rc2\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/broonie/regulator:\n  regulator: core: Return error in get optional stub\n  regulator: pbias: Convert to use regmap helper functions\n  regulator: pbias: Fix is_enabled callback implementation\n"
    },
    {
      "commit": "ff1e5b447e465da75dfdd0a4f4070cfef86a4155",
      "tree": "fdcf56b18824f8746f115165e72abe61a1fecca2",
      "parents": [
        "92891ed6b1fdb49655f9a071ef2880a567807375",
        "201f9e4dd5ecdefa70d4cf94750a667e2382d13f"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Thu Apr 24 12:01:05 2014 -0700"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Thu Apr 24 12:01:05 2014 -0700"
      },
      "message": "Merge tag \u0027spi-v3.15-rc2\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/broonie/spi\n\nPull spi fixes from Mark Brown:\n \"A few driver specific fixes here:\n\n   - SH HSPI was dealing with its clocks incorrectly which meant it\n     didn\u0027t work on some SoCs, fixing this also requires a small fix to\n     one of the SoC clock trees to avoid breaking existing users.\n   - The SiRF driver appears to have had several quality problems, it\u0027s\n     fairly new and not widely used so this isn\u0027t too worrying.\n   - A brute force fix for excessive locking in the Atmel driver, it\n     needs further investigation but this deals with the immediate\n     issue.\n   - A build fix for the Blackfin driver\"\n\n* tag \u0027spi-v3.15-rc2\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/broonie/spi:\n  spi: atmel: Fix scheduling while atomic bug\n  spi: sh-hspi: Do not specifically request shyway_clk clock\n  ARM: shmobile: r8a7778: Use clks as MSTP007 parent\n  spi: sirf: make GPIO chipselect function work well\n  spi: sirf: set SPI controller in RISC IO chipselect mode\n  spi: sirf: correct TXFIFO empty interrupt status bit\n  spi: bfin5xx: fix build error\n"
    },
    {
      "commit": "f66abe92ce498672ac44d550f51a526597c731c4",
      "tree": "d1938ba51ba9853b5c408984e1cd6f21bb2beca0",
      "parents": [
        "a798c10faf62a505d24e5f6213fbaf904a39623f"
      ],
      "author": {
        "name": "Rafael J. Wysocki",
        "email": "rafael.j.wysocki@intel.com",
        "time": "Thu Apr 24 19:27:49 2014 +0200"
      },
      "committer": {
        "name": "Rafael J. Wysocki",
        "email": "rafael.j.wysocki@intel.com",
        "time": "Thu Apr 24 19:27:49 2014 +0200"
      },
      "message": "ACPI / notify: Do not block unknown type notifications in root handler\n\nCommit 1a699476e258 \"ACPI / hotplug / PCI: Hotplug notifications from\nacpi_bus_notify()\" changed the root notify handler, acpi_bus_notify(),\nto block unknown type norifications, but it overlooked the fact that\nthey might be propagated to drivers via the -\u003enotify() callback.\n\nFix the problem by allowing drivers to receive unknown type\nnotifications via -\u003enotify() as before.\n\nFixes: 1a699476e258 (ACPI / hotplug / PCI: Hotplug notifications from acpi_bus_notify())\nReported-and-tested-by: Mantas Mikulėnas \u003cgrawity@gmail.com\u003e\nReported-and-tested-by: Sitsofe Wheeler \u003csitsofe@yahoo.com\u003e\nSigned-off-by: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\n"
    },
    {
      "commit": "92891ed6b1fdb49655f9a071ef2880a567807375",
      "tree": "f86be60d63d93815a2efb04d0ff3aa65c90cde8f",
      "parents": [
        "fdd324aa5fcaf3cc82f71bde51bf884436c9a986",
        "59f0f119e85cfb173db518328b55efbac4087c9f"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Thu Apr 24 09:58:57 2014 -0700"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Thu Apr 24 09:58:57 2014 -0700"
      },
      "message": "Merge branch \u0027fixes_for_v3.15\u0027 of git://git.linaro.org/people/mszyprowski/linux-dma-mapping\n\nPull dma-mapping fix from Marek Szyprowski:\n \"A small fix for dma-mapping subsystem for ARM\"\n\n* \u0027fixes_for_v3.15\u0027 of git://git.linaro.org/people/mszyprowski/linux-dma-mapping:\n  arm: dma-mapping: Fix mapping size value\n"
    },
    {
      "commit": "fdd324aa5fcaf3cc82f71bde51bf884436c9a986",
      "tree": "5c247d7cd214c5f80cfdc4478f0ea2e33fc5de49",
      "parents": [
        "db725c88c74669603239a757761a2aecc390f55d",
        "8a4aeec8d2d6a3edeffbdfae451cdf05cbf0fefd"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Thu Apr 24 09:57:02 2014 -0700"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Thu Apr 24 09:57:02 2014 -0700"
      },
      "message": "Merge branch \u0027for-3.15-fixes\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tj/libata\n\nPull libata fixes from Tejun Heo:\n \"Dan updated tag allocation to accomodate devices which choke when tags\n  jump back and forth.  Quite a few ahci MSI related fixes.  A couple\n  config dependency fixes and other misc fixes\"\n\n* \u0027for-3.15-fixes\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tj/libata:\n  libata/ahci: accommodate tag ordered controllers\n  ahci: Do not receive interrupts sent by dummy ports\n  ahci: Use pci_enable_msi_exact() instead of pci_enable_msi_range()\n  ahci: Ensure \"MSI Revert to Single Message\" mode is not enforced\n  ahci: do not request irq for dummy port\n  pata_samsung_cf: fix ata_host_activate() failure handling\n  pata_arasan_cf: fix ata_host_activate() failure handling\n  ata: fix i.MX AHCI driver dependencies\n  pata_at91: fix ata_host_activate() failure handling\n  libata: Update queued trim blacklist for M5x0 drives\n  libata: make AHCI_XGENE depend on PHY_XGENE\n"
    },
    {
      "commit": "cf2e0a73ca9ad376825c013ebaa145608abc27d7",
      "tree": "b6a4709dca0c7f1be56f183ad4a251b8fb7630bb",
      "parents": [
        "d08b80373cbb76c9b485b60d49fd3ba82abdf77c"
      ],
      "author": {
        "name": "Punit Agrawal",
        "email": "punit.agrawal@arm.com",
        "time": "Wed Mar 19 12:43:25 2014 +0000"
      },
      "committer": {
        "name": "Pawel Moll",
        "email": "pawel.moll@arm.com",
        "time": "Thu Apr 24 17:20:50 2014 +0100"
      },
      "message": "ARM: vexpress/TC2: Convert OPP voltage to uV before storing\n\nThe SPC stores voltage in mV while the code assumes it was returning\nuV. Convert the returned voltage to uV before storing. Also fix the\ncomment depicting voltage to uV.\n\nSigned-off-by: Punit Agrawal \u003cpunit.agrawal@arm.com\u003e\nReviewed-by: Sudeep Holla \u003csudeep.holla@arm.com\u003e\nSigned-off-by: Pawel Moll \u003cpawel.moll@arm.com\u003e\n"
    },
    {
      "commit": "d08b80373cbb76c9b485b60d49fd3ba82abdf77c",
      "tree": "79e58dc5a01b1a87719b374d59730d85f52fc149",
      "parents": [
        "bb6dd5757c12e9f3f25da971f1ea45c777ae79ab"
      ],
      "author": {
        "name": "Pawel Moll",
        "email": "pawel.moll@arm.com",
        "time": "Thu Apr 24 17:19:30 2014 +0100"
      },
      "committer": {
        "name": "Pawel Moll",
        "email": "pawel.moll@arm.com",
        "time": "Thu Apr 24 17:20:50 2014 +0100"
      },
      "message": "power/reset: vexpress: Fix restart/power off operation\n\nThe restart/power off implementation in the vexpress driver\nused to obtain the config function when necessary. This was\nwrong in two respects:\n\n1. It required memory allocation with disabled interrupts\n(it worked, but lockdep - when enabled - reported warnings).\n\n2. Used jiffies-based timeout, while jiffies are not running\nat this stage of system shutdown (therefore a config\ntransaction error - if happened - would have never be reported).\n\nFixed by pre-allocating the config function per device\nand using mdelay for timeout.\n\nSigned-off-by: Pawel Moll \u003cpawel.moll@arm.com\u003e\n"
    },
    {
      "commit": "9ef1af9ea28c23d0eaed97f7f5142788b6cf570a",
      "tree": "1c1c017be7711c7ea7245f963ef88775a31b5007",
      "parents": [
        "9ba71705706aa83bcd7f9b74ae2d167da934c951"
      ],
      "author": {
        "name": "Stephen Warren",
        "email": "swarren@nvidia.com",
        "time": "Tue Apr 01 14:13:17 2014 -0600"
      },
      "committer": {
        "name": "Arnd Bergmann",
        "email": "arnd@arndb.de",
        "time": "Thu Apr 24 15:37:08 2014 +0200"
      },
      "message": "dt: tegra: remove non-existent clock IDs\n\nThe Tegra124 clock DT binding currently provides 3 clocks that don\u0027t\nactually exist; 2 for NAND and one for UART5/UARTE. Delete these. While\nthis is technically an incompatible DT ABI change, nothing could have\nused these clock IDs for anything practical, since the HW doesn\u0027t exist.\n\nCc: \u003cstable@vger.kernel.org\u003e\nSigned-off-by: Stephen Warren \u003cswarren@nvidia.com\u003e\nSigned-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\n"
    },
    {
      "commit": "9ba71705706aa83bcd7f9b74ae2d167da934c951",
      "tree": "41fbdeb711470e09665044bd483e85691fad1cc3",
      "parents": [
        "862f0eea38409ff0d7f226c1245b787e3f0e2607"
      ],
      "author": {
        "name": "Stephen Warren",
        "email": "swarren@nvidia.com",
        "time": "Tue Apr 01 14:13:16 2014 -0600"
      },
      "committer": {
        "name": "Arnd Bergmann",
        "email": "arnd@arndb.de",
        "time": "Thu Apr 24 15:36:50 2014 +0200"
      },
      "message": "clk: tegra: remove non-existent clocks\n\nThe Tegra124 clock driver currently provides 3 clocks that don\u0027t actually\nexist; 2 for NAND and one for UART5/UARTE. Delete these.\n\nCc: \u003cstable@vger.kernel.org\u003e\nSigned-off-by: Stephen Warren \u003cswarren@nvidia.com\u003e\nSigned-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\n"
    },
    {
      "commit": "862f0eea38409ff0d7f226c1245b787e3f0e2607",
      "tree": "fd2b541bbd2f0c19aa3fae75c0e129452d2cc85d",
      "parents": [
        "f126776a2114e453ff48dc882603bae2d82bfc2f"
      ],
      "author": {
        "name": "Stephen Warren",
        "email": "swarren@nvidia.com",
        "time": "Tue Apr 01 14:13:15 2014 -0600"
      },
      "committer": {
        "name": "Arnd Bergmann",
        "email": "arnd@arndb.de",
        "time": "Thu Apr 24 15:36:40 2014 +0200"
      },
      "message": "ARM: tegra: remove UART5/UARTE from tegra124.dtsi\n\nTegra124 only has 4 UARTs. Parts of the documentation hint at a fifth\nUART, but this appears to be left-over from earlier SoC documentation.\nRemove the non-existent DT node for UART5.\n\nCc: \u003cstable@vger.kernel.org\u003e\nSigned-off-by: Stephen Warren \u003cswarren@nvidia.com\u003e\nSigned-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\n"
    },
    {
      "commit": "f126776a2114e453ff48dc882603bae2d82bfc2f",
      "tree": "e3a7122d4ac727ea93b6a70a38737709a0c48c2b",
      "parents": [
        "072c8b3fe2017c2cb342d6fe6b57063ab48f8bad",
        "fb677ef70b65e22cd4401d31b700a8b4041efae1"
      ],
      "author": {
        "name": "Arnd Bergmann",
        "email": "arnd@arndb.de",
        "time": "Thu Apr 24 15:35:58 2014 +0200"
      },
      "committer": {
        "name": "Arnd Bergmann",
        "email": "arnd@arndb.de",
        "time": "Thu Apr 24 15:35:58 2014 +0200"
      },
      "message": "Merge tag \u0027omap-for-v3.15/fixes-v2-signed\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tmlind/linux-omap into fixes\n\nFixes for omaps, mostly to fix some GPMC, DSS and USB issues for\ndevice tree based booting. And turns out BeagleBoard xM A/B\nneeds it\u0027s own minimal dts in addition to the related u-boot\nchanges. Also few minor documentation and typo fixes are merged\nto get them out of the way.\n\n* tag \u0027omap-for-v3.15/fixes-v2-signed\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tmlind/linux-omap:\n  ARM: OMAP2+: Fix GPMC remap for devices using an offset\n  ARM: OMAP2+: Fix oops for GPMC free\n  ARM: dts: Add support for the BeagleBoard xM A/B\n  ARM: dts: Grammar /that will/it will/\n  ARM: dts: Grammar /is uses/ is used/\n  ARM: OMAP2+: Fix config name for USB3 PHY\n  ARM: dts: am335x: update USB DT references\n  ARM: dts: OMAP2+: remove uses of obsolete gpmc,device-nand\n  ARM: AM335X: EVM: fix pinmux documentation in devicetree\n  ARM: OMAP2+: N900: remove omapdss init for DT boot\n  ARM: dts: dra7xx-clocks: Correct mcasp2_ahclkx_mux bit-shift\n  ARM: dts: omap5: Add clocks to USB3 PHY node\n  ARM: OMAP2+: hwmod: fix missing braces in _init()\n  ARM: AM43xx: fix dpll init in bypass mode\n  ARM: OMAP3: hwmod data: Correct clock domains for USB modules\n  ARM: OMAP3: PM: remove access to PRM_VOLTCTRL register\n\nSigned-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\n"
    },
    {
      "commit": "072c8b3fe2017c2cb342d6fe6b57063ab48f8bad",
      "tree": "e45adf076cd223501821e4cc3e8d76899475722e",
      "parents": [
        "6e1ae422c35d7b2274351dcf170fedd799f73288",
        "1fb4e407b520a2344b92a680c75672c7900aedf8"
      ],
      "author": {
        "name": "Arnd Bergmann",
        "email": "arnd@arndb.de",
        "time": "Thu Apr 24 15:35:14 2014 +0200"
      },
      "committer": {
        "name": "Arnd Bergmann",
        "email": "arnd@arndb.de",
        "time": "Thu Apr 24 15:35:14 2014 +0200"
      },
      "message": "Merge tag \u0027ux500-defconfig-for-arm-soc\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/linusw/linux-stericsson into fixes\n\nThis is a patch set for some ST-Ericsson devices:\n- Updates the Ux500 (U8500) defconfig\n- Selects PARTITION_ADVANCED for Ux500 and U300\n- Configure in IIO sensor drivers for the Ux500\n- Configure in the CW1200 WLAN chip for the Ux500\n\n* tag \u0027ux500-defconfig-for-arm-soc\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/linusw/linux-stericsson:\n  ARM: ux500: configure for CW1200 WLAN chip\n  ARM: ux500: configure in sensors\n  ARM: u300: u300_defconfig: Enable PARTITION_ADVANCED\n  ARM: ux500: u8500_defconfig: Enable PARTITION_ADVANCED\n  ARM: ux500: update defconfig\n\nSigned-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\n"
    },
    {
      "commit": "6e1ae422c35d7b2274351dcf170fedd799f73288",
      "tree": "6cd4e6f83060f0ed5b021f1d1ab2bc0ca805b56b",
      "parents": [
        "8a52a116d1c3e9ee93e3886eb87a419e32b22cd5",
        "d5dda0381f826326071b027f288372f682c24d55"
      ],
      "author": {
        "name": "Arnd Bergmann",
        "email": "arnd@arndb.de",
        "time": "Thu Apr 24 15:34:34 2014 +0200"
      },
      "committer": {
        "name": "Arnd Bergmann",
        "email": "arnd@arndb.de",
        "time": "Thu Apr 24 15:34:34 2014 +0200"
      },
      "message": "Merge tag \u0027renesas-fixes-for-v3.15\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/horms/renesas into fixes\n\nRenesas ARM Based SoC Fixes Updates for v3.15\n\nr8a7791 (R-Car M2) based koelsch board\n* Correct renesas,gpios to renesas,groups in sd[012] pfc\n\n8a7790 (R-Car H2) based lager board\n* Correct SND_SOC_DAIFMT_CBx_CFx flags\n\nr8a7740 (R-Mobile A1) SoC\n* Drop address cells from GIC node\n\nr8a7740 (R-Mobile A1) based Armadillo800 EVA board\n* Correct SND_SOC_DAIFMT_CBx_CFx flags\n\nsh73a0 (SH-Mobile AG5) SoC\n* Drop address cells from GIC node\n\n* tag \u0027renesas-fixes-for-v3.15\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/horms/renesas:\n  ARM: shmobile: koelsch: correct renesas,gpios to renesas,groups in sd[012] pfc\n  ARM: shmobile: r8a7740: drop address cells from GIC node\n  ARM: shmobile: sh73a0: drop address cells from GIC node\n  ARM: shmobile: armadillo800eva: fixup SND_SOC_DAIFMT_CBx_CFx flags\n  ARM: shmobile: lager: fixup SND_SOC_DAIFMT_CBx_CFx flags\n\nSigned-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\n"
    },
    {
      "commit": "8a52a116d1c3e9ee93e3886eb87a419e32b22cd5",
      "tree": "5fcba73bc5e7aafcc521d52082b3d2ec981af0b8",
      "parents": [
        "59209c9a34f1e57922055110a476f9bc56c3fc99",
        "b08eed0c4d7d70eae2f4f1ff518cc33643722a07"
      ],
      "author": {
        "name": "Arnd Bergmann",
        "email": "arnd@arndb.de",
        "time": "Thu Apr 24 15:34:00 2014 +0200"
      },
      "committer": {
        "name": "Arnd Bergmann",
        "email": "arnd@arndb.de",
        "time": "Thu Apr 24 15:34:00 2014 +0200"
      },
      "message": "Merge tag \u0027renesas-dt5-for-v3.15\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/horms/renesas into fixes\n\nFifth Round of Renesas ARM Based SoC DT Updates for v3.15\n\nCorrect renesas,groups in SDHI nodes of for r8a7790 (R-Car H2) based\nLager board.\n\n* tag \u0027renesas-dt5-for-v3.15\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/horms/renesas:\n  ARM: shmobile: lager: correct renesas,gpios to renesas,groups in sd[02] pfc\n\nSigned-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\n"
    },
    {
      "commit": "59209c9a34f1e57922055110a476f9bc56c3fc99",
      "tree": "337923f7f77c47fdfca014833a0a1b76ea92a5a9",
      "parents": [
        "9dbb7e24514f33144e7bdbe66aae84adc04b9cc8",
        "5c4348c1f959234664e4667f428cf82b3b8a7c2c"
      ],
      "author": {
        "name": "Arnd Bergmann",
        "email": "arnd@arndb.de",
        "time": "Thu Apr 24 15:33:31 2014 +0200"
      },
      "committer": {
        "name": "Arnd Bergmann",
        "email": "arnd@arndb.de",
        "time": "Thu Apr 24 15:33:31 2014 +0200"
      },
      "message": "Merge tag \u0027v3.15-rockchip-fixes1\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/mmind/linux-rockchip into fixes\n\nFixing uart-rx pull settings and a copy\u0027n\u0027paste error in a smp message\n\n* tag \u0027v3.15-rockchip-fixes1\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/mmind/linux-rockchip:\n  ARM: rockchip: fix copy\u0027n\u0027paste error in smp error messages\n  ARM: rockchip: rk3188: enable pull-ups on UART RX pins\n\nSigned-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\n"
    },
    {
      "commit": "9dbb7e24514f33144e7bdbe66aae84adc04b9cc8",
      "tree": "5f8ceb7473bb2fe986b9857ac49d1536eceb6e07",
      "parents": [
        "191bcd8120eb59c707e33ff159699a57db796eda",
        "4591b13289b54fb5cbce84ee170f7390c576ef8f"
      ],
      "author": {
        "name": "Arnd Bergmann",
        "email": "arnd@arndb.de",
        "time": "Thu Apr 24 15:30:53 2014 +0200"
      },
      "committer": {
        "name": "Arnd Bergmann",
        "email": "arnd@arndb.de",
        "time": "Thu Apr 24 15:30:53 2014 +0200"
      },
      "message": "Merge tag \u0027imx-fixes-3.15\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/shawnguo/linux into fixes\n\nThe i.MX fixes for 3.15:\n - A couple of dts changes for the fallout of imx-drm binding update\n - Parent DI clocks to video PLL for better HDMI support\n - PCIe interrupt mapping and GIC node fixes\n - A series of edmqmx6 board fixes\n - Other small and random fixes on imx5 and imx6 dts\n\n* tag \u0027imx-fixes-3.15\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/shawnguo/linux:\n  ARM: i.MX6: ipu_di_sel clocks can set parent rates\n  ARM: imx6q: clk: Parent DI clocks to video PLL via di_pre_sel\n  ARM: dts: imx: add required #clock-cells for fixed-clock\n  ARM: dts: vybrid: drop address and size cells from GIC node\n  ARM: dts: imx6sl-evk: Add an entry for MX6SL_PAD_ECSPI1_SS0__GPIO4_IO11\n  ARM: dts: imx53: fix apparent copy/paste error\n  ARM: dts: imx6q-gw5xxx: remove dead \u0027crtcs\u0027 property\n  ARM: dts: imx53-tx53: add IPU DI ports and endpoints\n  ARM: dts: imx6: edmqmx6: add second STMPE\n  ARM: dts: imx6: edmqmx6: USB H1 only supports host mode\n  ARM: dts: imx6: edmqmx6: Do not use the OTG switch as VBUS regulator\n  ARM: dts: imx6: edmqmx6: Fix usbotg id pin\n  ARM: dt: microsom: don\u0027t set bit 7 for ethernet mux settings\n  ARM: imx6q-clk: parent lvds_gate from lvds_sel\n  ARM: dts: imx: drop invalid size and address cells properties\n  ARM: dts: mx5: fix wrong stmpe-ts bindings\n  ARM: dts: imx53-m53evk: Fix memory region description\n  ARM: dts: imx53-qsb-common: Fix memory region description\n  ARM: dts: imx6: add PCIe interrupt mapping properties\n\nSigned-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\n"
    },
    {
      "commit": "191bcd8120eb59c707e33ff159699a57db796eda",
      "tree": "459e836162cd669eb58d15f3a83e17a3a165f4c7",
      "parents": [
        "cab4d50389c6b23fc3d7184d794856186bdc6651",
        "33faf20b8ebe2770f2d7b52796f5f35eeb87ab6f"
      ],
      "author": {
        "name": "Arnd Bergmann",
        "email": "arnd@arndb.de",
        "time": "Thu Apr 24 15:30:16 2014 +0200"
      },
      "committer": {
        "name": "Arnd Bergmann",
        "email": "arnd@arndb.de",
        "time": "Thu Apr 24 15:30:16 2014 +0200"
      },
      "message": "Merge tag \u0027mvebu-dt-fixes-non-crit-3.15\u0027 of git://git.infradead.org/linux-mvebu into fixes\n\nmvebu DT fixes-non-critical (for v3.15-rc1)\n\n - kirkwood\n    - add some missing vendor prefixes to keep checkpatch happy\n\n - mvebu\n    - add clock ref to mdio node on 370/XP/38x\n\n* tag \u0027mvebu-dt-fixes-non-crit-3.15\u0027 of git://git.infradead.org/linux-mvebu:\n  ARM: mvebu: ensure the mdio node has a clock reference on Armada 38x\n  ARM: mvebu: ensure the mdio node has a clock reference on Armada 370/XP\n  ARM: Kirkwood: DT: Add missing vendor prefix\n  ARM: Kirkwood: Fix Atmel vendor prefix\n\nSigned-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\n"
    },
    {
      "commit": "cab4d50389c6b23fc3d7184d794856186bdc6651",
      "tree": "6b70b43f478f8cfc16086c8646acd54dbd008fc9",
      "parents": [
        "af81c08c281bbc1dfc7b1fbf85f240fb7072724e"
      ],
      "author": {
        "name": "Paul Bolle",
        "email": "pebolle@tiscali.nl",
        "time": "Mon Apr 07 10:52:23 2014 -0600"
      },
      "committer": {
        "name": "Arnd Bergmann",
        "email": "arnd@arndb.de",
        "time": "Thu Apr 24 15:29:12 2014 +0200"
      },
      "message": "ARM: tegra: remove TEGRA_EMC_SCALING_ENABLE\n\nCommit a7cbe92cef27 (\"ARM: tegra: remove tegra EMC scaling driver\")\nremoved the only user of TEGRA_EMC_SCALING_ENABLE. Remove its Kconfig\nentry too.\n\nSigned-off-by: Paul Bolle \u003cpebolle@tiscali.nl\u003e\nSigned-off-by: Stephen Warren \u003cswarren@nvidia.com\u003e\nSigned-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\n"
    },
    {
      "commit": "af81c08c281bbc1dfc7b1fbf85f240fb7072724e",
      "tree": "0da120cad2b0452eccb06164f0f09c219bf05293",
      "parents": [
        "6d0add405a6665e743fddf9c117fd6778f7a428e"
      ],
      "author": {
        "name": "Domenico Andreoli",
        "email": "domenico.andreoli@linux.com",
        "time": "Mon Apr 14 11:55:01 2014 +0200"
      },
      "committer": {
        "name": "Arnd Bergmann",
        "email": "arnd@arndb.de",
        "time": "Thu Apr 24 15:28:47 2014 +0200"
      },
      "message": "ARM: Tidy up DTB Makefile entries\n\nFew things were out of order:\n\n- removed ARCH_BCM2835 duplicate\n- shuffled ARCH_BCM_5301X, ARCH_U8500 and ARCH_U300 around so to keep the\n  list sorted\n\nCc: Arnd Bergmann \u003carnd@arndb.de\u003e\nCc: Olof Johansson \u003colof@lixom.net\u003e\nSigned-by: Domenico Andreoli \u003cdomenico.andreoli@linux.com\u003e\n\nSigned-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\n"
    },
    {
      "commit": "6d0add405a6665e743fddf9c117fd6778f7a428e",
      "tree": "59956366974feaee0e6edddbbd8d0bf9c49f6994",
      "parents": [
        "1be5f69216a20b3ed7ef8b0392ca97385bdca959"
      ],
      "author": {
        "name": "Rob Herring",
        "email": "robh@kernel.org",
        "time": "Wed Apr 16 08:42:13 2014 -0500"
      },
      "committer": {
        "name": "Arnd Bergmann",
        "email": "arnd@arndb.de",
        "time": "Thu Apr 24 15:28:20 2014 +0200"
      },
      "message": "ARM: fix missing CLKSRC_OF on multi-platform\n\nIn commit ddb902cc34593e (ARM: centralize common multi-platform kconfig\noptions), CLKSRC_OF was removed from some platforms, but not added to\nARCH_MULTIPLATFORM. Fix this.\n\nReported-by: Lauri Hintsala \u003clauri.hintsala@bluegiga.com\u003e\nSigned-off-by: Rob Herring \u003crobh@kernel.org\u003e\nSigned-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\n"
    },
    {
      "commit": "1be5f69216a20b3ed7ef8b0392ca97385bdca959",
      "tree": "f89786c410565cc6671ab4fe21cd2c43bfdf5533",
      "parents": [
        "c02b50e90be9f41d6802049a1a08246e9eb1a22c"
      ],
      "author": {
        "name": "Alex Elder",
        "email": "elder@linaro.org",
        "time": "Wed Apr 09 09:01:12 2014 -0500"
      },
      "committer": {
        "name": "Arnd Bergmann",
        "email": "arnd@arndb.de",
        "time": "Thu Apr 24 15:28:00 2014 +0200"
      },
      "message": "ARM: spear: add __init to spear_clocksource_init()\n\nI get a build warning because spear_clocksource_init() calls\nclocksource_mmio_init(), but it doesn\u0027t have an __init annotation.\nFix that.\n\nSigned-off-by: Alex Elder \u003celder@linaro.org\u003e\nAcked-by: Viresh Kumar \u003cviresh.kumar@linaro.org\u003e\nSigned-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\n"
    },
    {
      "commit": "c02b50e90be9f41d6802049a1a08246e9eb1a22c",
      "tree": "1433a0c2c3c5535c5e893746890a6e54ee28f47f",
      "parents": [
        "c9eaa447e77efe77b7fa4c953bd62de8297fd6c5"
      ],
      "author": {
        "name": "Andrea Adami",
        "email": "andrea.adami@gmail.com",
        "time": "Tue Apr 08 14:53:44 2014 +0200"
      },
      "committer": {
        "name": "Arnd Bergmann",
        "email": "arnd@arndb.de",
        "time": "Thu Apr 24 15:26:39 2014 +0200"
      },
      "message": "ARM: pxa: hx4700.h: include \"irqs.h\" for PXA_NR_BUILTIN_GPIO\n\nhx4700 needs the same fix as in\n9705e74671f0e4f994d86b00cecf441917c64a66\n\"ARM: pxa: fix various compilation problems\"\n\nFix build errors. Initial one is:\n/linux/arch/arm/mach-pxa/include/mach/hx4700.h:18:32: error:\n \u0027PXA_NR_BUILTIN_GPIO\u0027 undeclared here (not in a function)\n|  #define HX4700_ASIC3_GPIO_BASE PXA_NR_BUILTIN_GPIO\n\nCc: stable@vger.kernel.org # v3.13+\nSigned-off-by: Andrea Adami \u003candrea.adami@gmail.com\u003e\nSigned-off-by: Linus Walleij \u003clinus.walleij@linaro.org\u003e\nSigned-off-by: Arnd Bergmann \u003carnd@arndb.de\u003e\n"
    },
    {
      "commit": "f75d72309192e52ef9a3efb390b1c4f408c142df",
      "tree": "7ba34fb0c598fca268d0db9e9f3c3585e62c565f",
      "parents": [
        "b2e5411ee26bff1deefd0b2c7beec9c133632e65"
      ],
      "author": {
        "name": "Guenter Roeck",
        "email": "linux@roeck-us.net",
        "time": "Mon Apr 21 10:38:08 2014 -0700"
      },
      "committer": {
        "name": "Guenter Roeck",
        "email": "linux@roeck-us.net",
        "time": "Thu Apr 24 06:09:43 2014 -0700"
      },
      "message": "hwmon: (ltc2945) Don\u0027t crash the kernel unnecessarily\n\nAn implementation error should not crash the kernel if it is avoidable.\nReplace BUG() with WARN_ONCE().\n\nSigned-off-by: Guenter Roeck \u003clinux@roeck-us.net\u003e\n"
    },
    {
      "commit": "b2e5411ee26bff1deefd0b2c7beec9c133632e65",
      "tree": "fca94d153e37e215cd54e223c5f37c3e60c8ef8a",
      "parents": [
        "52feaca5d685c07a4bc9812d16d84f5f7991bfe7"
      ],
      "author": {
        "name": "Pawel Moll",
        "email": "pawel.moll@arm.com",
        "time": "Wed Apr 23 18:27:05 2014 +0100"
      },
      "committer": {
        "name": "Guenter Roeck",
        "email": "linux@roeck-us.net",
        "time": "Thu Apr 24 06:02:10 2014 -0700"
      },
      "message": "hwmon: (vexpress) Avoid creating non-existing attributes\n\nThe \u0027label\u0027 attribute was always created but returned -ENOENT\nif there is no label and such behaviour is undefined from\nlibsensors\u0027 point of view.\n\nFixed by providing is_visible method in the attributes group,\nso the attribute is not created at all when unnecessary.\n\nReported-by: Guenter Roeck \u003clinux@roeck-us.net\u003e\nSigned-off-by: Pawel Moll \u003cpawel.moll@arm.com\u003e\nSigned-off-by: Guenter Roeck \u003clinux@roeck-us.net\u003e\n"
    },
    {
      "commit": "52feaca5d685c07a4bc9812d16d84f5f7991bfe7",
      "tree": "e08ee3ccbcc7b74c2f6bd5fb9c16509017e26df4",
      "parents": [
        "a798c10faf62a505d24e5f6213fbaf904a39623f"
      ],
      "author": {
        "name": "Pawel Moll",
        "email": "pawel.moll@arm.com",
        "time": "Wed Apr 23 18:27:04 2014 +0100"
      },
      "committer": {
        "name": "Guenter Roeck",
        "email": "linux@roeck-us.net",
        "time": "Thu Apr 24 06:01:59 2014 -0700"
      },
      "message": "hwmon: (vexpress) Use legal hwmon device names\n\nThe driver used to directly us a DT \u0027compatible\u0027 property for\nthe \u0027name\u0027 attribute of the hwmon devices. Unfortunately it\ncontains \u0027-\u0027 which is illegal in this context. It messes up\nlibsensors and thus every application using it.\n\nFixed by providing equivalent (and simpler) name strings.\n\nReported-by: Guenter Roeck \u003clinux@roeck-us.net\u003e\nSigned-off-by: Pawel Moll \u003cpawel.moll@arm.com\u003e\nSigned-off-by: Guenter Roeck \u003clinux@roeck-us.net\u003e\n"
    },
    {
      "commit": "bb6dd5757c12e9f3f25da971f1ea45c777ae79ab",
      "tree": "dee08962b88b6ec4a7ae5840e304ac3908aed08c",
      "parents": [
        "6b4ed8b00e93bd31f24a25f59ed8d1b808d0cc00"
      ],
      "author": {
        "name": "Heinrich Schuchardt",
        "email": "xypron.glpk@gmx.de",
        "time": "Mon Apr 14 08:42:01 2014 +0100"
      },
      "committer": {
        "name": "Pawel Moll",
        "email": "pawel.moll@arm.com",
        "time": "Thu Apr 24 11:41:12 2014 +0100"
      },
      "message": "arm/mach-vexpress: array accessed out of bounds\n\ndcscb_allcpus_mask is an array of size 2.\n\nThe index variable cluster has to be checked against this limit\nbefore accessing the array.\n\nSigned-off-by: Heinrich Schuchardt \u003cxypron.glpk@gmx.de\u003e\nAcked-by: Nicolas Pitre \u003cnico@linaro.org\u003e\nSigned-off-by: Pawel Moll \u003cpawel.moll@arm.com\u003e\n"
    },
    {
      "commit": "6b4ed8b00e93bd31f24a25f59ed8d1b808d0cc00",
      "tree": "2570de1eabab44273210e6d3e3bff4749f5840dd",
      "parents": [
        "a798c10faf62a505d24e5f6213fbaf904a39623f"
      ],
      "author": {
        "name": "Dan Carpenter",
        "email": "dan.carpenter@oracle.com",
        "time": "Thu Nov 07 08:08:44 2013 +0000"
      },
      "committer": {
        "name": "Pawel Moll",
        "email": "pawel.moll@arm.com",
        "time": "Thu Apr 24 11:39:06 2014 +0100"
      },
      "message": "clk: vexpress: NULL dereference on error path\n\nIf the allocation fails then we dereference the NULL in the error path.\nJust return directly.\n\nFixes: ed27ff1db869 (\u0027clk: Versatile Express clock generators (\"osc\") driver\u0027)\nSigned-off-by: Dan Carpenter \u003cdan.carpenter@oracle.com\u003e\nSigned-off-by: Pawel Moll \u003cpawel.moll@arm.com\u003e\n"
    },
    {
      "commit": "cb171f7abb9a1a250fb41d088b81799f75bd1357",
      "tree": "f644f4e236082876cc3940f2ac2c703a8c4638f6",
      "parents": [
        "a798c10faf62a505d24e5f6213fbaf904a39623f"
      ],
      "author": {
        "name": "Bjorn Helgaas",
        "email": "bhelgaas@google.com",
        "time": "Wed Apr 23 10:21:06 2014 -0600"
      },
      "committer": {
        "name": "Rafael J. Wysocki",
        "email": "rafael.j.wysocki@intel.com",
        "time": "Thu Apr 24 02:39:40 2014 +0200"
      },
      "message": "PNP: Work around BIOS defects in Intel MCH area reporting\n\nWork around BIOSes that don\u0027t report the entire Intel MCH area.\n\nMCHBAR is not an architected PCI BAR, so MCH space is usually reported as a\nPNP0C02 resource.  The MCH space was once 16KB, but is 32KB in newer parts.\nSome BIOSes still report a PNP0C02 resource that is only 16KB, which means\nthe rest of the MCH space is consumed but unreported.\n\nThis can cause resource map sanity check warnings or (theoretically) a\ndevice conflict if we assigned the unreported space to another device.\n\nThe Intel perf event uncore driver tripped over this when it claimed the\nMCH region:\n\n  resource map sanity check conflict: 0xfed10000 0xfed15fff 0xfed10000 0xfed13fff pnp 00:01\n  Info: mapping multiple BARs. Your kernel is fine.\n\nTo prevent this, if we find a PNP0C02 resource that covers part of the MCH\nspace, extend it to cover the entire space.\n\nReferences: http://lkml.kernel.org/r/20140224162400.GE16457@pd.tnic\nReported-and-tested-by: Borislav Petkov \u003cbp@alien8.de\u003e\nSigned-off-by: Bjorn Helgaas \u003cbhelgaas@google.com\u003e\nAcked-by: Borislav Petkov \u003cbp@suse.de\u003e\nAcked-by: Stephane Eranian \u003ceranian@google.com\u003e\nSigned-off-by: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\n"
    },
    {
      "commit": "db725c88c74669603239a757761a2aecc390f55d",
      "tree": "7e549a6d102bd6044cc87082a7c9cb2b4c7951ef",
      "parents": [
        "1aae31c8306e5f1bdeafd87b2cd9e3f0df3709e5",
        "556d3f7f4d791cae54fd24ef28296e666f4c96a6"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Wed Apr 23 16:48:56 2014 -0700"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Wed Apr 23 16:48:56 2014 -0700"
      },
      "message": "Merge branch \u0027fixes\u0027 of git://ftp.arm.linux.org.uk/~rmk/linux-arm\n\nPull ARM fixes from Russell King:\n \"Various fixes and post-merge window updates.  Included here are:\n   - ensure Kconfig things which should be sorted remain sorted\n   - fix three big-endian bugs which crept in during the last merge\n     window\n   - add the renameat2 syscall\n   - fix big.LITTLE switcher initialisation checks\n   - fix kdump vmcore for LPAE kernels\"\n\n* \u0027fixes\u0027 of git://ftp.arm.linux.org.uk/~rmk/linux-arm:\n  ARM: add renameat2 syscall\n  ARM: keep arch/arm/Kconfig and arch/arm/mm/Kconfig select entries sorted\n  ARM: 8033/1: fix big endian __pv_phys_pfn_offset size related issue\n  ARM: 8032/1: bL_switcher: fix validation check before its activation\n  ARM: 8030/1: ARM : kdump : add arch_crash_save_vmcoreinfo\n  ARM: 8027/1: fix do_div() bug in big-endian systems\n  ARM: 8026/1: Fix emulation of multiply accumulate instructions\n  ARM: 8024/1: Keep DEBUG_UART_{PHYS,VIRT} entries sorted\n"
    },
    {
      "commit": "2704f807f9498054b8153002bafa3e818079e9a5",
      "tree": "a4cca20ab339fcc3832a9c864bf02fd57ae2f2e7",
      "parents": [
        "8b425aa193439a2b60a988a9230f1de753e36902"
      ],
      "author": {
        "name": "H Hartley Sweeten",
        "email": "hsweeten@visionengravers.com",
        "time": "Fri Mar 28 09:20:58 2014 -0700"
      },
      "committer": {
        "name": "Greg Kroah-Hartman",
        "email": "gregkh@linuxfoundation.org",
        "time": "Wed Apr 23 16:20:15 2014 -0700"
      },
      "message": "staging: comedi: usbdux: bug fix for accessing \u0027ao_chanlist\u0027 in private data\n\nIn usbdux_ao_cmd(), the channels for the command are transfered from the\ncmd-\u003echanlist and stored in the private data \u0027ao_chanlist\u0027. The channel\nnumbers are bit-shifted when stored so that they become the \"command\"\nthat is transfered to the device. The channel to command conversion\nresults in the \u0027ao_chanlist\u0027 having these values for the channels:\n\n  channel 0 -\u003e ao_chanlist \u003d 0x00\n  channel 1 -\u003e ao_chanlist \u003d 0x40\n  channel 2 -\u003e ao_chanlist \u003d 0x80\n  channel 3 -\u003e ao_chanlist \u003d 0xc0\n\nThe problem is, the usbduxsub_ao_isoc_irq() function uses the \u0027chan\u0027 value\nfrom \u0027ao_chanlist\u0027 to access the \u0027ao_readback\u0027 array in the private data.\nSo instead of accessing the array as 0, 1, 2, 3, it accesses it as 0x00,\n0x40, 0x80, 0xc0.\n\nFix this by storing the raw channel number in \u0027ao_chanlist\u0027 and doing the\nbit-shift when creating the command.\n\nFixes: a998a3db530bff80 \"staging: comedi: usbdux: cleanup the private data \u0027outBuffer\u0027\"\nCc: stable \u003cstable@vger.kernel.org\u003e # 3.12\nSigned-off-by: H Hartley Sweeten \u003chsweeten@visionengravers.com\u003e\nReviewed-by: Ian Abbott \u003cabbotti@mev.co.uk\u003e\nAcked-by: Bernd Porr \u003cmail@berndporr.me.uk\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\n"
    }
  ],
  "next": "cff2fce58b2b0f59089e7edcdc38803d65057b9f"
}