1. 15 1月, 2011 1 次提交
  2. 14 1月, 2011 5 次提交
  3. 13 1月, 2011 7 次提交
  4. 12 1月, 2011 27 次提交
    • L
      eca0bdd3
    • L
      tools: create power/x86/x86_energy_perf_policy · d5532ee7
      Len Brown 提交于
      MSR_IA32_ENERGY_PERF_BIAS first became available on Westmere Xeon.
      It is implemented in all Sandy Bridge processors -- mobile, desktop and server.
      It is expected to become increasingly important in subsequent generations.
      
      x86_energy_perf_policy is a user-space utility to set the
      hardware energy vs performance policy hint in the processor.
      Most systems would benefit from "x86_energy_perf_policy normal"
      at system startup, as the hardware default is maximum performance
      at the expense of energy efficiency.
      
      See x86_energy_perf_policy.8 man page for more information.
      
      Background:
      
      Linux-2.6.36 added "epb" to /proc/cpuinfo to indicate
      if an x86 processor supports MSR_IA32_ENERGY_PERF_BIAS,
      without actually modifying the MSR.
      
      In March, 2010, Venkatesh Pallipadi proposed a small driver
      that programmed MSR_IA32_ENERGY_PERF_BIAS, based on
      the cpufreq governor in use.  It also offered
      a boot-time cmdline option to override.
      http://lkml.org/lkml/2010/3/4/457
      But hiding the hardware policy behind the
      governor choice was deemed "kinda icky".
      
      In June, 2010, I proposed a generic user/kernel API to
      generalize the power/performance policy trade-off.
      "RFC: /sys/power/policy_preference"
      http://lkml.org/lkml/2010/6/16/399
      That is my preference for implementing this capability,
      but I received no support on the list.
      
      So in September, 2010, I sent x86_energy_perf_policy.c to LKML,
      a user-space utility that scribbles directly to the MSR.
      http://lkml.org/lkml/2010/9/28/246
      
      Here is that same utility, after responding to some review feedback,
      to live in tools/power/, where it is easily found.
      Signed-off-by: NLen Brown <len.brown@intel.com>
      d5532ee7
    • L
      tools: create power/x86/turbostat · 103a8fea
      Len Brown 提交于
      turbostat is a Linux tool to observe proper operation
      of Intel(R) Turbo Boost Technology.
      
      turbostat displays the actual processor frequency
      on x86 processors that include APERF and MPERF MSRs.
      
      Note that turbostat is of limited utility on Linux
      kernels 2.6.29 and older, as acpi_cpufreq cleared
      APERF/MPERF up through that release.
      
      On Intel Core i3/i5/i7 (Nehalem) and newer processors,
      turbostat also displays residency in idle power saving states,
      which are necessary for diagnosing any cpuidle issues
      that may have an effect on turbo-mode.
      
      See the turbostat.8 man page for example usage.
      Signed-off-by: NLen Brown <len.brown@intel.com>
      103a8fea
    • D
      i915/gtt: fix ordering causing DMAR errors on object teardown. · d15eda5c
      Dave Airlie 提交于
      Previous to the last GTT rework we always rewrote the GTT then unmapped the
      object, somehow this got reversed in the rework in 2.6.37-rc5 timeframe.
      
      This fix needs to go to stable in an alternate form since the code changed.
      
      This fixes DMAR reports on my Ironlake HP2540p.
      Signed-off-by: NDave Airlie <airlied@redhat.com>
      d15eda5c
    • D
      i915/gtt: fix ordering issues with status setup and DMAR · a46f3108
      Dave Airlie 提交于
      This code was setting up the status page before setting the DMAR-is-on-bit,
      so we were getting DMAR errors on the status page. Reverse the two bits
      of init code to the correct result.
      Signed-off-by: NDave Airlie <airlied@redhat.com>
      a46f3108
    • D
      Merge branch 'drm-intel-fixes' of ssh://master.kernel.org/pub/scm/linux/kernel/git/ickle/drm-intel · 784fe39f
      Dave Airlie 提交于
      * 'drm-intel-fixes' of ssh://master.kernel.org/pub/scm/linux/kernel/git/ickle/drm-intel: (37 commits)
        drm/i915/execbuffer: Reorder binding of objects to favour restrictions
        drm/i915: If we hit OOM when allocating GTT pages, clear the aperture
        drm/i915/evict: Ensure we completely cleanup on failure
        drm/i915/execbuffer: Correctly clear the current object list upon EFAULT
        drm/i915/debugfs: Show all objects in the gtt
        drm/i915: Record AGP memory type upon error
        drm/i915: Periodically flush the active lists and requests
        drm/i915/gtt: Unmap the PCI pages after unbinding them from the GTT
        drm/i915: Record the error batchbuffer on each ring
        drm/i915: Include TLB miss overhead for computing WM
        drm/i915: Propagate error from flushing the ring
        drm/i915: detect & report PCH display error interrupts
        drm/i915: cleanup rc6 code
        drm/i915: fix rc6 enabling around suspend/resume
        drm/i915: re-enable rc6 support for Ironlake+
        drm/i915: Make the ring IMR handling private
        drm/i915/ringbuffer: Simplify the ring irq refcounting
        drm/i915/debugfs: Show the per-ring IMR
        drm/i915: Mask USER interrupts on gen6 (until required)
        drm/i915: Handle ringbuffer stalls when flushing
        ...
      784fe39f
    • L
      Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-2.6 · 4162cf64
      Linus Torvalds 提交于
      * git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-2.6: (67 commits)
        cxgb4vf: recover from failure in cxgb4vf_open()
        netfilter: ebtables: make broute table work again
        netfilter: fix race in conntrack between dump_table and destroy
        ah: reload pointers to skb data after calling skb_cow_data()
        ah: update maximum truncated ICV length
        xfrm: check trunc_len in XFRMA_ALG_AUTH_TRUNC
        ehea: Increase the skb array usage
        net/fec: remove config FEC2 as it's used nowhere
        pcnet_cs: add new_id
        tcp: disallow bind() to reuse addr/port
        net/r8169: Update the function of parsing firmware
        net: ppp: use {get,put}_unaligned_be{16,32}
        CAIF: Fix IPv6 support in receive path for GPRS/3G
        arp: allow to invalidate specific ARP entries
        net_sched: factorize qdisc stats handling
        mlx4: Call alloc_etherdev to allocate RX and TX queues
        net: Add alloc_netdev_mqs function
        caif: don't set connection request param size before copying data
        cxgb4vf: fix mailbox data/control coherency domain race
        qlcnic: change module parameter permissions
        ...
      4162cf64
    • L
      Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/sparc-2.6 · fb7b5a95
      Linus Torvalds 提交于
      * git://git.kernel.org/pub/scm/linux/kernel/git/davem/sparc-2.6:
        sparc64: Fix bootup regression due to perf init ordering.
      fb7b5a95
    • L
      Merge branch 'next' of git://git.kernel.org/pub/scm/linux/kernel/git/benh/powerpc · 5a62f995
      Linus Torvalds 提交于
      * 'next' of git://git.kernel.org/pub/scm/linux/kernel/git/benh/powerpc: (72 commits)
        powerpc/pseries: Fix build of topology stuff without CONFIG_NUMA
        powerpc/pseries: Fix VPHN build errors on non-SMP systems
        powerpc/83xx: add mpc8308_p1m DMA controller device-tree node
        powerpc/83xx: add DMA controller to mpc8308 device-tree node
        powerpc/512x: try to free dma descriptors in case of allocation failure
        powerpc/512x: add MPC8308 dma support
        powerpc/512x: fix the hanged dma transfer issue
        powerpc/512x: scatter/gather dma fix
        powerpc/powermac: Make auto-loading of therm_pm72 possible
        of/address: Use propper endianess in get_flags
        powerpc/pci: Use printf extension %pR for struct resource
        powerpc: Remove unnecessary casts of void ptr
        powerpc: Disable VPHN polling during a suspend operation
        powerpc/pseries: Poll VPA for topology changes and update NUMA maps
        powerpc: iommu: Add device name to iommu error printks
        powerpc: Record vma->phys_addr in ioremap()
        powerpc: Update compat_arch_ptrace
        powerpc: Fix PPC_PTRACE_SETHWDEBUG on PPC_BOOK3S
        powerpc/time: printk time stamp init not correct
        powerpc: Minor cleanups for machdep.h
        ...
      5a62f995
    • L
      Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/roland/infiniband · f1d6d6cd
      Linus Torvalds 提交于
      * 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/roland/infiniband: (42 commits)
        IB/qib: Fix refcount leak in lkey/rkey validation
        IB/qib: Improve SERDES tunning on QMH boards
        IB/qib: Unnecessary delayed completions on RC connection
        IB/qib: Issue pre-emptive NAKs on eager buffer overflow
        IB/qib: RDMA lkey/rkey validation is inefficient for large MRs
        IB/qib: Change QPN increment
        IB/qib: Add fix missing from earlier patch
        IB/qib: Change receive queue/QPN selection
        IB/qib: Fix interrupt mitigation
        IB/qib: Avoid duplicate writes to the rcv head register
        IB/qib: Add a few new SERDES tunings
        IB/qib: Reset packet list after freeing
        IB/qib: New SERDES init routine and improvements to SI quality
        IB/qib: Clear WAIT_SEND flags when setting QP to error state
        IB/qib: Fix context allocation with multiple HCAs
        IB/qib: Fix multi-Florida HCA host panic on reboot
        IB/qib: Handle transitions from ACTIVE_DEFERRED to ACTIVE better
        IB/qib: UD send with immediate receive completion has wrong size
        IB/qib: Set port physical state even if other fields are invalid
        IB/qib: Generate completion callback on errors
        ...
      f1d6d6cd
    • B
    • C
      cxgb4vf: recover from failure in cxgb4vf_open() · 343a8d13
      Casey Leedom 提交于
      If the Link Start fails in cxgb4vf_open(), we need to back out any state
      that we've built up ...
      Signed-off-by: NCasey Leedom <leedom@chelsio.com>
      Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      343a8d13
    • D
      60dbb011
    • L
      Merge branch 'nfs-for-2.6.38' of git://git.linux-nfs.org/projects/trondmy/nfs-2.6 · b9d919a4
      Linus Torvalds 提交于
      * 'nfs-for-2.6.38' of git://git.linux-nfs.org/projects/trondmy/nfs-2.6: (89 commits)
        NFS fix the setting of exchange id flag
        NFS: Don't use vm_map_ram() in readdir
        NFSv4: Ensure continued open and lockowner name uniqueness
        NFS: Move cl_delegations to the nfs_server struct
        NFS: Introduce nfs_detach_delegations()
        NFS: Move cl_state_owners and related fields to the nfs_server struct
        NFS: Allow walking nfs_client.cl_superblocks list outside client.c
        pnfs: layout roc code
        pnfs: update nfs4_callback_recallany to handle layouts
        pnfs: add CB_LAYOUTRECALL handling
        pnfs: CB_LAYOUTRECALL xdr code
        pnfs: change lo refcounting to atomic_t
        pnfs: check that partial LAYOUTGET return is ignored
        pnfs: add layout to client list before sending rpc
        pnfs: serialize LAYOUTGET(openstateid)
        pnfs: layoutget rpc code cleanup
        pnfs: change how lsegs are removed from layout list
        pnfs: change layout state seqlock to a spinlock
        pnfs: add prefix to struct pnfs_layout_hdr fields
        pnfs: add prefix to struct pnfs_layout_segment fields
        ...
      b9d919a4
    • F
      netfilter: ebtables: make broute table work again · 2f46e079
      Florian Westphal 提交于
      broute table init hook sets up the "br_should_route_hook" pointer,
      which then gets called from br_input.
      
      commit a386f990
      (bridge: add proper RCU annotation to should_route_hook)
      introduced a typedef, and then changed this to:
      
      br_should_route_hook_t *rhook;
      [..]
      rhook = rcu_dereference(br_should_route_hook);
      if (*rhook(skb))
      
      problem is that "br_should_route_hook" contains the address of the function,
      so calling *rhook() results in kernel panic.
      Signed-off-by: NFlorian Westphal <fw@strlen.de>
      Acked-by: NEric Dumazet <eric.dumazet@gmail.com>
      Signed-off-by: NPablo Neira Ayuso <pablo@netfilter.org>
      2f46e079
    • C
      drm/i915/execbuffer: Reorder binding of objects to favour restrictions · 6fe4f140
      Chris Wilson 提交于
      As the mappable portion of the aperture is always a small subset at the
      start of the GTT, it is allocated preferentially by drm_mm. This is
      useful in case we ever need to map an object later. However, if you have
      a large object that can consume the entire mappable region of the
      GTT this prevents the batchbuffer from fitting and so causing an error.
      Instead allocate all those that require a mapping up front in order to
      improve the likelihood of finding sufficient space to bind them.
      Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
      6fe4f140
    • C
      drm/i915: If we hit OOM when allocating GTT pages, clear the aperture · 809b6334
      Chris Wilson 提交于
      Rather than evicting an object at random, which is unlikely to alleviate
      the memory pressure sufficient to allow us to continue, zap the entire
      aperture. That should give the system long enough to recover and reap
      some pages from the evicted objects, forestalling the allocation error
      for the new object.
      Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
      809b6334
    • C
      drm/i915/evict: Ensure we completely cleanup on failure · 092de6f2
      Chris Wilson 提交于
      ... and not leave the objects in a inconsistent state.
      Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
      Cc: stable@kernel.org
      092de6f2
    • C
      drm/i915/execbuffer: Correctly clear the current object list upon EFAULT · 36cf1742
      Chris Wilson 提交于
      Before releasing the lock in order to copy the relocation list from user
      pages, we need to drop all the object references as another thread may
      usurp and execute another batchbuffer before we reacquire the lock.
      However, the code was buggy and failed to clear the list...
      Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
      Cc: stable@kernel.org
      36cf1742
    • S
      netfilter: fix race in conntrack between dump_table and destroy · 13ee6ac5
      Stephen Hemminger 提交于
      The netlink interface to dump the connection tracking table has a race
      when entries are deleted at the same time. A customer reported a crash
      and the backtrace showed thatctnetlink_dump_table was running while a
      conntrack entry was being destroyed.
      (see https://bugzilla.vyatta.com/show_bug.cgi?id=6402).
      
      According to RCU documentation, when using hlist_nulls the reader
      must handle the case of seeing a deleted entry and not proceed
      further down the linked list.  The old code would continue
      which caused the scan to walk into the free list.
      
      This patch uses locking (rather than RCU) for this operation which
      is guaranteed safe, and no longer requires getting reference while
      doing dump operation.
      Signed-off-by: NStephen Hemminger <shemminger@vyatta.com>
      Signed-off-by: NPablo Neira Ayuso <pablo@netfilter.org>
      13ee6ac5
    • L
      Merge branch 'for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jack/linux-udf-2.6 · 7c955fca
      Linus Torvalds 提交于
      * 'for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jack/linux-udf-2.6:
        UDF: Close small mem leak in udf_find_entry()
        udf: Fix directory corruption after extent merging
        udf: Protect udf_file_aio_write from possible races
        udf: Remove unnecessary bkl usages
        udf: Use of s_alloc_mutex to serialize udf_relocate_blocks() execution
        udf: Replace bkl with the UDF_I(inode)->i_data_sem for protect udf_inode_info struct
        udf: Remove BKL from free space counting functions
        udf: Call udf_add_free_space() for more blocks at once in udf_free_blocks()
        udf: Remove BKL from udf_put_super() and udf_remount_fs()
        udf: Protect default inode credentials by rwlock
        udf: Protect all modifications of LVID with s_alloc_mutex
        udf: Move handling of uniqueID into a helper function and protect it by a s_alloc_mutex
        udf: Remove BKL from udf_update_inode
        udf: Convert UDF_SB(sb)->s_flags to use bitops
        fs/udf: Add printf format/argument verification
        fs/udf: Use vzalloc
      
      (Evil merge: this also removes the BKL dependency from the Kconfig file)
      7c955fca
    • L
      Merge branch 'for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tytso/ext4 · e9688f6a
      Linus Torvalds 提交于
      * 'for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tytso/ext4: (44 commits)
        ext4: fix trimming starting with block 0 with small blocksize
        ext4: revert buggy trim overflow patch
        ext4: don't pass entire map to check_eofblocks_fl
        ext4: fix memory leak in ext4_free_branches
        ext4: remove ext4_mb_return_to_preallocation()
        ext4: flush the i_completed_io_list during ext4_truncate
        ext4: add error checking to calls to ext4_handle_dirty_metadata()
        ext4: fix trimming of a single group
        ext4: fix uninitialized variable in ext4_register_li_request
        ext4: dynamically allocate the jbd2_inode in ext4_inode_info as necessary
        ext4: drop i_state_flags on architectures with 64-bit longs
        ext4: reorder ext4_inode_info structure elements to remove unneeded padding
        ext4: drop ec_type from the ext4_ext_cache structure
        ext4: use ext4_lblk_t instead of sector_t for logical blocks
        ext4: replace i_delalloc_reserved_flag with EXT4_STATE_DELALLOC_RESERVED
        ext4: fix 32bit overflow in ext4_ext_find_goal()
        ext4: add more error checks to ext4_mkdir()
        ext4: ext4_ext_migrate should use NULL not 0
        ext4: Use ext4_error_file() to print the pathname to the corrupted inode
        ext4: use IS_ERR() to check for errors in ext4_error_file
        ...
      e9688f6a
    • L
      Merge branch 'for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jack/linux-fs-2.6 · 40c73abb
      Linus Torvalds 提交于
      * 'for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jack/linux-fs-2.6:
        ext2: Resolve 'dereferencing pointer to incomplete type' when enabling EXT2_XATTR_DEBUG
        ext3: Remove redundant unlikely()
        ext2: Remove redundant unlikely()
        ext3: speed up file creates by optimizing rec_len functions
        ext2: speed up file creates by optimizing rec_len functions
        ext3: Add more journal error check
        ext3: Add journal error check in resize.c
        quota: Use %pV and __attribute__((format (printf in __quota_error and fix fallout
        ext3: Add FITRIM handling
        ext3: Add batched discard support for ext3
        ext3: Add journal error check into ext3_rename()
        ext3: Use search_dirblock() in ext3_dx_find_entry()
        ext3: Avoid uninitialized memory references with a corrupted htree directory
        ext3: Return error code from generic_check_addressable
        ext3: Add journal error check into ext3_delete_entry()
        ext3: Add error check in ext3_mkdir()
        fs/ext3/super.c: Use printf extension %pV
        fs/ext2/super.c: Use printf extension %pV
        ext3: don't update sb journal_devnum when RO dev
      40c73abb
    • L
      Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/ericvh/v9fs · 0945f352
      Linus Torvalds 提交于
      * 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/ericvh/v9fs:
        fs/9p: Don't set dentry->d_op in create routines
        fs/9p: fix spelling typo
        fs/9p: TREADLINK bugfix
        net/9p: Use proper data types
        fs/9p: Simplify the .L create operation
        fs/9p: Move dotl inode operations into a seperate file
        fs/9p: fix menu presentation
        fs/9p: Fix the return error on default acl removal
        fs/9p: Remove unnecessary semicolons
      0945f352
    • C
      drm/i915/debugfs: Show all objects in the gtt · 08c18323
      Chris Wilson 提交于
      Useful for determining the layout.
      Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
      08c18323
    • C
      drm/i915: Record AGP memory type upon error · a779e5ab
      Chris Wilson 提交于
      Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
      a779e5ab
    • C
      drm/i915: Periodically flush the active lists and requests · 0a58705b
      Chris Wilson 提交于
      In order to retire active buffers whilst no client is active, we need to
      insert our own flush requests onto the ring.
      
      This is useful for servers that queue up some rendering and then go to
      sleep as it allows us to the complete processing of those requests,
      potentially making that memory available again much earlier.
      Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
      0a58705b