Permalink
Switch branches/tags
Commits on May 3, 2010
  1. Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel…

    …/git/sage/ceph-client
    
    * 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/sage/ceph-client:
      ceph: remove bad auth_x kmem_cache
      ceph: fix lockless caps check
      ceph: clear dir complete, invalidate dentry on replayed rename
      ceph: fix direct io truncate offset
      ceph: discard incoming messages with bad seq #
      ceph: fix seq counting for skipped messages
      ceph: add missing #includes
      ceph: fix leaked spinlock during mds reconnect
      ceph: print more useful version info on module load
      ceph: fix snap realm splits
      ceph: clear dir complete on d_move
    torvalds committed May 3, 2010
  2. ceph: remove bad auth_x kmem_cache

    It's useless, since our allocations are already a power of 2.  And it was
    allocated per-instance (not globally), which caused a name collision when
    we tried to mount a second file system with auth_x enabled.
    
    Signed-off-by: Sage Weil <sage@newdream.net>
    liewegas committed Apr 29, 2010
  3. ceph: fix lockless caps check

    The __ variant requires caller to hold i_lock.
    
    Signed-off-by: Sage Weil <sage@newdream.net>
    liewegas committed Apr 23, 2010
  4. ceph: clear dir complete, invalidate dentry on replayed rename

    If a rename operation is resent to the MDS following an MDS restart, the
    client does not get a full reply (containing the resulting metadata) back.
    In that case, a ceph_rename() needs to compensate by doing anything useful
    that fill_inode() would have, like d_move().
    
    It also needs to invalidate the dentry (to workaround the vfs_rename_dir()
    bug) and clear the dir complete flag, just like fill_trace().
    
    Signed-off-by: Sage Weil <sage@newdream.net>
    liewegas committed Apr 28, 2010
  5. ceph: fix direct io truncate offset

    truncate_inode_pages_range wants the end offset to align with the last byte
    in a page.
    
    Signed-off-by: Sage Weil <sage@newdream.net>
    liewegas committed Apr 22, 2010
  6. ceph: discard incoming messages with bad seq #

    We can get old message seq #'s after a tcp reconnect for stateful sessions
    (i.e., the MDS).  If we get a higher seq #, that is an error, and we
    shouldn't see any bad seq #'s for stateless (mon, osd) connections.
    
    Signed-off-by: Sage Weil <sage@newdream.net>
    liewegas committed Apr 22, 2010
  7. ceph: fix seq counting for skipped messages

    Increment in_seq even when the message is skipped for some reason.
    
    Signed-off-by: Sage Weil <sage@newdream.net>
    liewegas committed Apr 22, 2010
  8. ceph: add missing #includes

    Signed-off-by: Sage Weil <sage@newdream.net>
    liewegas committed Apr 20, 2010
  9. ceph: fix leaked spinlock during mds reconnect

    Signed-off-by: Sage Weil <sage@newdream.net>
    liewegas committed Apr 20, 2010
  10. ceph: print more useful version info on module load

    Decouple the client version from the server side.  Print relevant protocol
    and map version info instead.
    
    Signed-off-by: Sage Weil <sage@newdream.net>
    liewegas committed Apr 19, 2010
  11. ceph: fix snap realm splits

    The snap realm split was checking i_snap_realm, not the list_head, to
    determine if an inode belonged in the new realm.  The check always failed,
    which meant we always moved the inode, corrupting the old realm's list and
    causing various crashes.
    
    Also wait to release old realm reference to avoid possibility of use after
    free.
    
    Signed-off-by: Sage Weil <sage@newdream.net>
    liewegas committed Apr 19, 2010
  12. ceph: clear dir complete on d_move

    d_move() reorders the d_subdirs list, breaking the readdir result caching.
    Unless/until d_move preserves that ordering, clear CEPH_I_COMPLETE on
    rename.
    
    Signed-off-by: Sage Weil <sage@newdream.net>
    liewegas committed Apr 16, 2010
  13. Merge git://git.kernel.org/pub/scm/linux/kernel/git/wim/linux-2.6-wat…

    …chdog
    
    * git://git.kernel.org/pub/scm/linux/kernel/git/wim/linux-2.6-watchdog:
      watchdog: ep93xx_wdt.c fix default timout value in MODULE_PARM_DESC string.
    torvalds committed May 3, 2010
  14. nilfs2: fix sync silent failure

    As of 32a88aa, __sync_filesystem() will return 0 if s_bdi is not set.
    And nilfs does not set s_bdi anywhere.  I noticed this problem by the
    warning introduced by the recent commit 5129a46 ("Catch filesystem
    lacking s_bdi").
    
     WARNING: at fs/super.c:959 vfs_kern_mount+0xc5/0x14e()
     Hardware name: PowerEdge 2850
     Modules linked in: nilfs2 loop tpm_tis tpm tpm_bios video shpchp pci_hotplug output dcdbas
     Pid: 3773, comm: mount.nilfs2 Not tainted 2.6.34-rc6-debug #38
     Call Trace:
      [<c1028422>] warn_slowpath_common+0x60/0x90
      [<c102845f>] warn_slowpath_null+0xd/0x10
      [<c1095936>] vfs_kern_mount+0xc5/0x14e
      [<c1095a03>] do_kern_mount+0x32/0xbd
      [<c10a811e>] do_mount+0x671/0x6d0
      [<c1073794>] ? __get_free_pages+0x1f/0x21
      [<c10a684f>] ? copy_mount_options+0x2b/0xe2
      [<c107b634>] ? strndup_user+0x48/0x67
      [<c10a81de>] sys_mount+0x61/0x8f
      [<c100280c>] sysenter_do_call+0x12/0x32
    
    This ensures to set s_bdi for nilfs and fixes the sync silent failure.
    
    Signed-off-by: Ryusuke Konishi <konishi.ryusuke@lab.ntt.co.jp>
    Acked-by: Jens Axboe <jens.axboe@oracle.com>
    Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
    konis committed with torvalds May 3, 2010
  15. watchdog: ep93xx_wdt.c fix default timout value in MODULE_PARM_DESC s…

    …tring.
    
    The WATCHDOG_TIMEOUT macro does not exist. The default timeout value is WDT_TIMEOUT.
    Fix the MODULE_PARM_DESC so that the code can compile again.
    
    reported-by: Randy Dunlap <randy.dunlap@oracle.com>
    Signed-off-by: Wim Van Sebroeck <wim@iguana.be>
    Wim Van Sebroeck committed May 3, 2010
Commits on Apr 30, 2010
  1. Merge branch 'upstream' of git://ftp.linux-mips.org/pub/scm/upstream-…

    …linus
    
    * 'upstream' of git://ftp.linux-mips.org/pub/scm/upstream-linus: (27 commits)
      MIPS: Loongson 2F: Fix of problems introduced by -mfix-loongson2f-jump
      MIPS: Loongson-2F: Use CONFIG_CPU_JUMP_WORKAROUNDS to control workarounds.
      MIPS: Loongson 2F: Enable fixups of the latest binutils
      MIPS: Loongson: Add CPU_LOONGSON2F_WORKAROUNDS
      MIPS: Kconfig: Make Broadcom SoC support naming consistent
      MIPS: BCM63xx: Update defconfig
      MIPS: oprofile: Fix breakage when CONFIG_OPROFILE=m
      STAGING: octeon-ethernet: Use proper phy addresses for Movidis hardware.
      NET: mdio-octeon: Enable the hardware before using it.
      I2C: Fix section mismatch errors in i2c-octeon.c
      MIPS: Loongson: Fix LOONGSON_ADDRWIN_CFG macro.
      MIPS: Loongson: Fix phys_mem_access_prot() check
      MIPS: Loongson: Fix find_vga_mem_init()
      MIPS: Loongson: Fix typo in gdium mach type string.
      MIPS: Use CKSEG1ADDR for uncached handler
      MIPS: Check for accesses beyond the end of the PGD.
      MIPS: Use uasm_i_ds{r,l}l_safe() instead of uasm_i_ds{r,l}l() in tlbex.c
      MIPS: Add uasm_i_dsrl_safe() and uasm_i_dsll_safe() to uasm.
      MIPS: die() does not call die notifier chain
      MIPS: Swarm, Littlesur: Enable PATA platform driver.
      ...
    torvalds committed Apr 30, 2010
  2. MIPS: Loongson 2F: Fix of problems introduced by -mfix-loongson2f-jump

    The -mfix-loongson2f-jump option provided by latest CVS binutils have fixed
    the out-of-order issue of Loongson-2F described in chapter 15 of the
    Loongson2F User Manual [1, 2], but introduced some problems.
    
    The option changes all of the jump target to "addr & 0xcfffffff" through the
    at($1) register, but for the reboot address of Loongson 2F 0xbfc00000 this is
    wrong.  Avoids the problem via telling the assembler to not use the $at
    register.
    
    [1] Loongson2F User Manual (Chinese Version)
        http://www.loongson.cn/uploadfile/file/200808211
    [2] English Version of Chapter 15:
        http://groups.google.com.hk/group/loongson-dev/msg/e0d2e220958f10a6?dmode=source
    
    Reported-and-tested-by: Liu Shiwei <liushiwei@gmail.com>
    Signed-off-by: Wu Zhangjin <wuzhangjin@gmail.com>
    Cc: linux-mips <linux-mips@linux-mips.org>
    Patchwork: http://patchwork.linux-mips.org/patch/1109/
    Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
    lzufalcon committed with ralfbaechle Apr 10, 2010
  3. MIPS: Loongson-2F: Use CONFIG_CPU_JUMP_WORKAROUNDS to control workaro…

    …unds.
    
    Signed-off-by: Wu Zhangjin <wuzhangjin@gmail.com>
    Cc: linux-mips <linux-mips@linux-mips.org>
    Patchwork: http://patchwork.linux-mips.org/patch/1106/
    Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
    lzufalcon committed with ralfbaechle Apr 10, 2010
  4. MIPS: Loongson 2F: Enable fixups of the latest binutils

    With the  "Fixups of Loongson2F" patch [1] having been applied to binutils
    for binutils 2.20.1 we now can use it's time to enable the options provided
    by the patch to compile the kernel.
    
    Without these fixups, the system may hang if the erratum is triggered.
    
    For more information on these fixups please refer to the following
    references.
    
    [1] "Fixups of Loongson2F" patch for binutils(actually for gas)
        http://sourceware.org/ml/binutils/2009-11/msg00387.html
    [2] Chapter 15 of "Loongson2F User Manual"(Chinese Version)
        http://www.loongson.cn/uploadfile/file/200808211
    [3] Chapter 15 of the English version Loongson 2F User Manual
        http://groups.google.com.hk/group/loongson-dev/msg/e0d2e220958f10a6?dmode=source
    
    Signed-off-by: Zhang Le <r0bertz@gentoo.org>
    Signed-off-by: Wu Zhangjin <wuzhangjin@gmail.com>
    Cc: Zhang Le <r0bertz@gentoo.org>
    Cc: linux-mips <linux-mips@linux-mips.org>
    Patchwork: http://patchwork.linux-mips.org/patch/1106/
    Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
    lzufalcon committed with ralfbaechle Apr 10, 2010
  5. MIPS: Loongson: Add CPU_LOONGSON2F_WORKAROUNDS

    As documented in the  Loongson 2F User Manual [2, 3], the old Loongson2F
    series (2F01 / 2F02) have the NOP & JUMP issues which requires workarounds
    in the kernel and binutils.  This issue has been rectified in Loongson 2F
    series 2F03 so no workarounds needed.
    
    Now that the workarounds [1] adding the the -mfix-loongson2f-nop and
    -mfix-loongson2f-jump options have been comitted to the binutils the CVS
    repository), we can add the workarounds in the kernel.
    
    The workarounds have no significant side effect on the system but may
    decrease performance so we control them through a a new
    CPU_LOONGSON2F_WORKAROUNDS config option allowing the users to only enable
    it as necessary.
    
    [1] "Fixups of Loongson2F" patch for binutils(actually for gas)
        http://sourceware.org/ml/binutils/2009-11/msg00387.html
    [2] Chapter 15 of "Loongson2F User Manual"(Chinese Version)
        http://www.loongson.cn/uploadfile/file/200808211
    [3] English Version of the above chapter 15
        http://groups.google.com.hk/group/loongson-dev/msg/e0d2e220958f10a6?dmode=source
    
    Signed-off-by: Wu Zhangjin <wuzhangjin@gmail.com>
    Cc: linux-mips <linux-mips@linux-mips.org>
    Patchwork: http://patchwork.linux-mips.org/patch/1105/
    Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
    lzufalcon committed with ralfbaechle Apr 10, 2010
  6. MIPS: Kconfig: Make Broadcom SoC support naming consistent

    Signed-off-by: Florian Fainelli <ffainelli@freebox.fr>
    To: linux-mips@linux-mips.org
    Patchwork: http://patchwork.linux-mips.org/patch/1082/
    Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
    ffainelli committed with ralfbaechle Mar 25, 2010
  7. MIPS: BCM63xx: Update defconfig

    the defconfig was out-of-sync since 2.6.30-rc6, update it with the new
    symbols and enable BCM6338, 6345, wireless, b43 driver and LEDs support.
    
    Signed-off-by:  Fainelli <ffainelli@freebox.fr>
    To: linux-mips@linux-mips.org
    Patchwork: http://patchwork.linux-mips.org/patch/1081/
    Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
    ffainelli committed with ralfbaechle Mar 25, 2010
  8. MIPS: oprofile: Fix breakage when CONFIG_OPROFILE=m

    When the oprofile is compiled as a module do_IRQ() is not called in
    arch/mips/loongson/lemote-2f/irq.c due to a wrong #ifdef there.
    
    Signed-off-by: Wu Zhangjin <wuzhangjin@gmail.com>
    Cc: linux-mips@linux-mips.org
    Patchwork: http://patchwork.linux-mips.org/patch/1143/
    Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
    lzufalcon committed with ralfbaechle Apr 26, 2010
  9. STAGING: octeon-ethernet: Use proper phy addresses for Movidis hardware.

    Signed-off-by: David Daney <ddaney@caviumnetworks.com>
    To: linux-mips@linux-mips.org
    To: netdev@vger.kernel.org
    To: gregkh@suse.de
    Patchwork: http://patchwork.linux-mips.org/patch/1091/
    Acked-by: David S. Miller <davem@davemloft.net>
    Acked-by: Greg Kroah-Hartman <gregkh@suse.de>
    Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
    David Daney committed with ralfbaechle Apr 2, 2010
  10. NET: mdio-octeon: Enable the hardware before using it.

    In some cases the mdio bus is not enabled at the time of probing.
    This prevents anything from working, so we will enable it before
    trying to use it, and disable it when the driver is removed.
    
    Signed-off-by: David Daney <ddaney@caviumnetworks.com>
    To: linux-mips@linux-mips.org
    To: netdev@vger.kernel.org
    To: gregkh@suse.de
    Patchwork: http://patchwork.linux-mips.org/patch/1090/
    Acked-by: David S. Miller <davem@davemloft.net>
    Acked-by: Greg Kroah-Hartman <gregkh@suse.de>
    Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
    David Daney committed with ralfbaechle Apr 2, 2010
  11. I2C: Fix section mismatch errors in i2c-octeon.c

    Signed-off-by: David Daney <ddaney@caviumnetworks.com>
    To: linux-i2c@vger.kernel.org
    To: ben-linux@fluff.org
    To: khali@linux-fr.org
    Cc: linux-mips@linux-mips.org
    Cc: rade.bozic.ext@nsn.com
    Patchwork: http://patchwork.linux-mips.org/patch/1037/
    Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
    David Daney committed with ralfbaechle Mar 8, 2010
  12. MIPS: Loongson: Fix LOONGSON_ADDRWIN_CFG macro.

    There's a typo in the LOONGSON_ADDRWIN_CFG macro. The cpu window mmap
    register address should contain the destination parameters not the
    source one.  This has not been noticed because the code is only using
    source = destination.
    
    Signed-off-by: Arnaud Patard <apatard@mandriva.com>
    To: linux-mips@linux-mips.org
    Patchwork: http://patchwork.linux-mips.org/patch/1162/
    Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
    Arnaud Patard committed with ralfbaechle Apr 29, 2010
  13. MIPS: Loongson: Fix phys_mem_access_prot() check

    The check used to determine if uncached accelerated should be used or not
    is wrong. The parenthesis are misplaced and making the test fail.
    
    Signed-off-by: Arnaud Patard <apatard@mandriva.com>
    To: linux-mips@linux-mips.org
    Patchwork: http://patchwork.linux-mips.org/patch/1161/
    Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
    Arnaud Patard committed with ralfbaechle Apr 29, 2010
  14. MIPS: Loongson: Fix find_vga_mem_init()

    This allows to use all display device for instance DISPLAY_OTHER like SM501.
    
    Signed-off-by: Arnaud Patard <apatard@mandriva.com>
    To: linux-mips@linux-mips.org
    Patchwork: http://patchwork.linux-mips.org/patch/1160/
    Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
    Richard LIU committed with ralfbaechle Apr 29, 2010
  15. MIPS: Loongson: Fix typo in gdium mach type string.

    It's not "gidum" but "gdium".
    
    Signed-off-by: Arnaud Patard <apatard@mandriva.com>
    Cc: linux-mips@linux-mips.org
    Patchwork: http://patchwork.linux-mips.org/patch/1159/
    Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
    Arnaud Patard committed with ralfbaechle Apr 29, 2010
  16. MIPS: Use CKSEG1ADDR for uncached handler

    "MIPS: Calculate proper ebase value for 64-bit kernels"
    9af43ea080dd5d6c7b34f38261780e5dd43537bc (lmo) rsp.
    f6be75d (kernel.org) broke some 64-bit
    MIPS systems.
    
    Before this we were using XKPHYS/cached as ebase and computed the uncached
    xphsys/unchached address for that area. After that commit ebase became a
    32-bit compat address and convert does not work anymore.  We now should use
    CKSEG1 for this. CKSEG1ADDR does just that in 32-bit and 64-bit.
    
    Signed-off-by: Sebastian Andrzej Siewior <sebastian@breakpoint.cc>
    To: Ralf Baechle <ralf@linux-mips.org>
    Patchwork: http://patchwork.linux-mips.org/patch/1149/
    Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
    Sebastian Andrzej Siewior committed with ralfbaechle Apr 27, 2010
  17. MIPS: Check for accesses beyond the end of the PGD.

    For some combinations of PAGE_SIZE and vmbits, it is possible to have
    userspace access that are beyond what is covered by the PGD, but within
    vmbits.  Such an access would cause the TLB refill handler to load garbage
    values for PMD and PTE potentially giving userspace access to parts of the
    physical address space to which it is not entitled.
    
    In the TLB refill hot path, we add a single dsrl instruction so we can
    check if any bits outside of the range covered by the PGD are set.  In
    the vmalloc side we then separate the bad case from the normal vmalloc
    case and call tlb_do_page_fault_0 if warranted.  This slows us down a
    bit, but has the benefit of yielding deterministic behavior.
    
    [Ralf: Fixed build error for 32-bit kernels.]
    [Ralf: Folded lmo commit c8c0e22b2aa3982852b44279638ef37f9aa31b7d into this
     commit.]
    
    Signed-off-by: David Daney <ddaney@caviumnetworks.com>
    To: linux-mips@linux-mips.org
    Patchwork: http://patchwork.linux-mips.org/patch/1152/
    Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
    
    ---
    David Daney committed with ralfbaechle Apr 28, 2010
  18. MIPS: Use uasm_i_ds{r,l}l_safe() instead of uasm_i_ds{r,l}l() in tlbex.c

    This makes the code somewhat cleaner while reducing the risk of shift
    amount overflows when various page table related options are changed.
    
    Signed-off-by: David Daney <ddaney@caviumnetworks.com>
    To: linux-mips@linux-mips.org
    Patchwork: http://patchwork.linux-mips.org/patch/1154/
    Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
    David Daney committed with ralfbaechle Apr 28, 2010
  19. MIPS: Add uasm_i_dsrl_safe() and uasm_i_dsll_safe() to uasm.

    This allows us to clean up the code by not having to explicitly code
    checks for shift amounts greater than 32.
    
    Signed-off-by: David Daney <ddaney@caviumnetworks.com>
    To: linux-mips@linux-mips.org
    Patchwork: http://patchwork.linux-mips.org/patch/1153/
    Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
    David Daney committed with ralfbaechle Apr 28, 2010
  20. MIPS: die() does not call die notifier chain

    The MIPS implementation of die() forgets to call notify_die() and thus notifiers
    registered via register_die_notifier() are not called.  This results in kgdb not
    being activated on exceptions.
    
    The only subtlety is that notify_die declares its regs argument w/o const, so
    the const had to be removed from mips die() as well.
    
    [Ralf: Fixed build error for SGI IP22 and IP28 platforms.]
    
    Signed-off-by: Yury Polyanskiy <ypolyans@princeton.edu>
    Cc: linux-mips@linux-mips.org
    Patchworks: http://patchwork.linux-mips.org/patch/1142/
    Acked-by: Jason Wessel <jason.wessel@windriver.com>
    Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
    
    ---
    Yury Polyanskiy committed with ralfbaechle Apr 26, 2010