Skip to content

K
Kernel

openeuler / Kernel
1 年多 前同步成功

通知 8
Star 0
Fork 0
K
Kernel
切换分支/标签
  1. 04 2月, 2009 8 次提交
    • C
      Btrfs: Change btree locking to use explicit blocking points · b4ce94de
      由 Chris Mason 提交于 
      Most of the btrfs metadata operations can be protected by a spinlock,
but some operations still need to schedule.

So far, btrfs has been using a mutex along with a trylock loop,
most of the time it is able to avoid going for the full mutex, so
the trylock loop is a big performance gain.

This commit is step one for getting rid of the blocking locks entirely.
btrfs_tree_lock takes a spinlock, and the code explicitly switches
to a blocking lock when it starts an operation that can schedule.

We'll be able get rid of the blocking locks in smaller pieces over time.
Tracing allows us to find the most common cause of blocking, so we
can start with the hot spots first.

The basic idea is:

btrfs_tree_lock() returns with the spin lock held

btrfs_set_lock_blocking() sets the EXTENT_BUFFER_BLOCKING bit in
the extent buffer flags, and then drops the spin lock.  The buffer is
still considered locked by all of the btrfs code.

If btrfs_tree_lock gets the spinlock but finds the blocking bit set, it drops
the spin lock and waits on a wait queue for the blocking bit to go away.

Much of the code that needs to set the blocking bit finishes without actually
blocking a good percentage of the time.  So, an adaptive spin is still
used against the blocking bit to avoid very high context switch rates.

btrfs_clear_lock_blocking() clears the blocking bit and returns
with the spinlock held again.

btrfs_tree_unlock() can be called on either blocking or spinning locks,
it does the right thing based on the blocking bit.

ctree.c has a helper function to set/clear all the locked buffers in a
path as blocking.
Signed-off-by: NChris Mason <chris.mason@oracle.com>
      b4ce94de
    • C
      Btrfs: hash_lock is no longer needed · c487685d
      由 Chris Mason 提交于 
      Before metadata is written to disk, it is updated to reflect that writeout
has begun.  Once this update is done, the block must be cow'd before it
can be modified again.

This update was originally synchronized by using a per-fs spinlock.  Today
the buffers for the metadata blocks are locked before writeout begins,
and everyone that tests the flag has the buffer locked as well.

So, the per-fs spinlock (called hash_lock for no good reason) is no
longer required.
Signed-off-by: NChris Mason <chris.mason@oracle.com>
      c487685d
    • C
      Btrfs: disable leak debugging checks in extent_io.c · 3935127c
      由 Chris Mason 提交于 
      extent_io.c has debugging code to report and free leaked extent_state
and extent_buffer objects at rmmod time.  This helps track down
leaks and it saves you from rebooting just to properly remove the
kmem_cache object.

But, the code runs under a fairly expensive spinlock and the checks to
see if it is currently enabled are not entirely consistent.  Some use
#ifdef and some #if.

This changes everything to #if and disables the leak checking.
Signed-off-by: NChris Mason <chris.mason@oracle.com>
      3935127c
    • C
      Btrfs: sort references by byte number during btrfs_inc_ref · b7a9f29f
      由 Chris Mason 提交于 
      When a block goes through cow, we update the reference counts of
everything that block points to.  The internal pointers of the block
can be in just about any order, and it is likely to have clusters of
things that are close together and clusters of things that are not.

To help reduce the seeks that come with updating all of these reference
counts, sort them by byte number before actual updates are done.
Signed-off-by: NChris Mason <chris.mason@oracle.com>
      b7a9f29f
    • C
      Btrfs: async threads should try harder to find work · b51912c9
      由 Chris Mason 提交于 
      Tracing shows the delay between when an async thread goes to sleep
and when more work is added is often very short.  This commit adds
a little bit of delay and extra checking to the code right before
we schedule out.

It allows more work to be added to the worker
without requiring notifications from other procs.
Signed-off-by: NChris Mason <chris.mason@oracle.com>
      b51912c9
    • J
      Btrfs: selinux support · 0279b4cd
      由 Jim Owens 提交于 
      Add call to LSM security initialization and save
resulting security xattr for new inodes.

Add xattr support to symlink inode ops.

Set inode->i_op for existing special files.
Signed-off-by: Njim owens <jowens@hp.com>
      0279b4cd
    • C
      Btrfs: make btrfs acls selectable · bef62ef3
      由 Christian Hesse 提交于 
      This patch adds a menu entry to kconfig to enable acls for btrfs.
This allows you to enable FS_POSIX_ACL at kernel compile time.

(updated by Jeff Mahoney to make the changes in fs/btrfs/Kconfig instead)
Signed-off-by: NChristian Hesse <mail@earthworm.de>
Signed-off-by: NJeff Mahoney <jeffm@suse.com>
      bef62ef3
    • C
      Btrfs: Catch missed bios in the async bio submission thread · a6837051
      由 Chris Mason 提交于 
      The async bio submission thread was missing some bios that were
added after it had decided there was no work left to do.
Signed-off-by: NChris Mason <chris.mason@oracle.com>
      a6837051
  3. 29 1月, 2009 5 次提交
    • C
      Btrfs: fix readdir on 32 bit machines · 89f135d8
      由 Chris Mason 提交于 
      After btrfs_readdir has gone through all the directory items, it
sets the directory f_pos to the largest possible int.  This way
applications that mix readdir with creating new files don't
end up in an endless loop finding the new directory items as they go.

It was a workaround for a bug in git, but the assumption was that if git
could make this looping mistake than it would be a common problem.

The largest possible int chosen was INT_LIMIT(typeof(file->f_pos),
and it is possible for that to be a larger number than 32 bit glibc
expects to come out of readdir.

This patches switches that to INT_LIMIT(off_t), which should keep
applications happy on 32 and 64 bit machines.
Signed-off-by: NChris Mason <chris.mason@oracle.com>
      89f135d8
    • C
      Merge branch 'master' of git://git.kernel.org/pub/scm/linux/kernel/git/mason/btrfs-unstable · e4f722fa
      由 Chris Mason 提交于 
      Fix fs/btrfs/super.c conflict around #includes
      e4f722fa
    • L
      Linux 2.6.29-rc3 · 18e352e4
      由 Linus Torvalds 提交于
      18e352e4
    • L
      Merge branch 'merge' of git://git.kernel.org/pub/scm/linux/kernel/git/benh/powerpc · c4568d6c
      由 Linus Torvalds 提交于 
      * 'merge' of git://git.kernel.org/pub/scm/linux/kernel/git/benh/powerpc:
  powerpc/mm: Fix handling of _PAGE_COHERENT in BAT setup code
  powerpc/pseries: Correct VIO bus accounting problem in CMO env.
  powerpc: More printing warning fixes for the l64 to ll64 conversion
  powerpc: Remove arch/ppc cruft from Kconfig
  powerpc: Printing fix for l64 to ll64 conversion: phyp_dump.c
  powerpc/embedded6xx: Update defconfigs
  powerpc/8xx: Update defconfigs
  powerpc/86xx: Update defconfigs
  powerpc/83xx: Update defconfigs
  powerpc/85xx: Update defconfigs
  powerpc/mpc8313erdb: fix kernel panic because mdio device is not probed
  powerpc/4xx: Update multi-board PowerPC 4xx defconfigs
  powerpc/44x: Update PowerPC 44x defconfigs
  powerpc/40x: Update PowerPC 40x defconfigs
  powerpc/85xx: Fix typo in mpc8572ds dts
  powerpc/44x: Warp patches for the new NDFC driver
  powerpc/4xx: DTS: Add Add'l SDRAM0 Compatible and Interrupt Info
      c4568d6c
    • L
      Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/gerg/m68knommu · 78a768b6
      由 Linus Torvalds 提交于 
      * 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/gerg/m68knommu:
  m68knommu: fix 5329 ColdFire periphal addressing
  uclinux: add process name to allocation error message
  m68knommu: correct the mii calculations for 532x ColdFire FEC
  m68knommu: add ColdFire M532x to the FEC configuration options
  m68knommu: fix syscall restarting
  m68knommu: remove the obsolete and long unused comempci chip support
  m68knommu: remove the no longer used PCI support option
  m68knommu: remove obsolete and unused eLIA board
  m68knommu: set NO_DMA
  m68knommu: fix cache flushing for the 527x ColdFire processors
  m68knommu: fix ColdFire 5272 serial baud rates in mcf.c
  m68knommu: use one exist from execption
      78a768b6
  5. 28 1月, 2009 27 次提交