Skip to content

cloud-kernel
cloud-kernel

openanolis / cloud-kernel
1 年多 前同步成功

通知 161
Star 36
Fork 7
cloud-kernel
cloud-kernel
切换分支/标签
  1. 08 1月, 2011 21 次提交
  3. 07 1月, 2011 19 次提交
    • N
      fs: scale mntget/mntput · b3e19d92
      由 Nick Piggin 提交于 
      The problem that this patch aims to fix is vfsmount refcounting scalability.
We need to take a reference on the vfsmount for every successful path lookup,
which often go to the same mount point.

The fundamental difficulty is that a "simple" reference count can never be made
scalable, because any time a reference is dropped, we must check whether that
was the last reference. To do that requires communication with all other CPUs
that may have taken a reference count.

We can make refcounts more scalable in a couple of ways, involving keeping
distributed counters, and checking for the global-zero condition less
frequently.

- check the global sum once every interval (this will delay zero detection
  for some interval, so it's probably a showstopper for vfsmounts).

- keep a local count and only taking the global sum when local reaches 0 (this
  is difficult for vfsmounts, because we can't hold preempt off for the life of
  a reference, so a counter would need to be per-thread or tied strongly to a
  particular CPU which requires more locking).

- keep a local difference of increments and decrements, which allows us to sum
  the total difference and hence find the refcount when summing all CPUs. Then,
  keep a single integer "long" refcount for slow and long lasting references,
  and only take the global sum of local counters when the long refcount is 0.

This last scheme is what I implemented here. Attached mounts and process root
and working directory references are "long" references, and everything else is
a short reference.

This allows scalable vfsmount references during path walking over mounted
subtrees and unattached (lazy umounted) mounts with processes still running
in them.

This results in one fewer atomic op in the fastpath: mntget is now just a
per-CPU inc, rather than an atomic inc; and mntput just requires a spinlock
and non-atomic decrement in the common case. However code is otherwise bigger
and heavier, so single threaded performance is basically a wash.
Signed-off-by: NNick Piggin <npiggin@kernel.dk>
      b3e19d92
    • N
      fs: provide rcu-walk aware permission i_ops · b74c79e9
      由 Nick Piggin 提交于 
      Signed-off-by: NNick Piggin <npiggin@kernel.dk>
      b74c79e9
    • N
      fs: rcu-walk aware d_revalidate method · 34286d66
      由 Nick Piggin 提交于 
      Require filesystems be aware of .d_revalidate being called in rcu-walk
mode (nd->flags & LOOKUP_RCU). For now do a simple push down, returning
-ECHILD from all implementations.
Signed-off-by: NNick Piggin <npiggin@kernel.dk>
      34286d66
    • N
      fs: dcache reduce branches in lookup path · fb045adb
      由 Nick Piggin 提交于 
      Reduce some branches and memory accesses in dcache lookup by adding dentry
flags to indicate common d_ops are set, rather than having to check them.
This saves a pointer memory access (dentry->d_op) in common path lookup
situations, and saves another pointer load and branch in cases where we
have d_op but not the particular operation.

Patched with:

git grep -E '[.>]([[:space:]])*d_op([[:space:]])*=' | xargs sed -e 's/\([^\t ]*\)->d_op = \(.*\);/d_set_d_op(\1, \2);/' -e 's/\([^\t ]*\)\.d_op = \(.*\);/d_set_d_op(\&\1, \2);/' -i
Signed-off-by: NNick Piggin <npiggin@kernel.dk>
      fb045adb
    • N
      fs: icache RCU free inodes · fa0d7e3d
      由 Nick Piggin 提交于 
      RCU free the struct inode. This will allow:

- Subsequent store-free path walking patch. The inode must be consulted for
  permissions when walking, so an RCU inode reference is a must.
- sb_inode_list_lock to be moved inside i_lock because sb list walkers who want
  to take i_lock no longer need to take sb_inode_list_lock to walk the list in
  the first place. This will simplify and optimize locking.
- Could remove some nested trylock loops in dcache code
- Could potentially simplify things a bit in VM land. Do not need to take the
  page lock to follow page->mapping.

The downsides of this is the performance cost of using RCU. In a simple
creat/unlink microbenchmark, performance drops by about 10% due to inability to
reuse cache-hot slab objects. As iterations increase and RCU freeing starts
kicking over, this increases to about 20%.

In cases where inode lifetimes are longer (ie. many inodes may be allocated
during the average life span of a single inode), a lot of this cache reuse is
not applicable, so the regression caused by this patch is smaller.

The cache-hot regression could largely be avoided by using SLAB_DESTROY_BY_RCU,
however this adds some complexity to list walking and store-free path walking,
so I prefer to implement this at a later date, if it is shown to be a win in
real situations. I haven't found a regression in any non-micro benchmark so I
doubt it will be a problem.
Signed-off-by: NNick Piggin <npiggin@kernel.dk>
      fa0d7e3d
    • N
      fs: dcache rationalise dget variants · dc0474be
      由 Nick Piggin 提交于 
      dget_locked was a shortcut to avoid the lazy lru manipulation when we already
held dcache_lock (lru manipulation was relatively cheap at that point).
However, how that the lru lock is an innermost one, we never hold it at any
caller, so the lock cost can now be avoided. We already have well working lazy
dcache LRU, so it should be fine to defer LRU manipulations to scan time.
Signed-off-by: NNick Piggin <npiggin@kernel.dk>
      dc0474be
    • N
      fs: dcache remove dcache_lock · b5c84bf6
      由 Nick Piggin 提交于 
      dcache_lock no longer protects anything. remove it.
Signed-off-by: NNick Piggin <npiggin@kernel.dk>
      b5c84bf6
    • N
      fs: Use rename lock and RCU for multi-step operations · 949854d0
      由 Nick Piggin 提交于 
      The remaining usages for dcache_lock is to allow atomic, multi-step read-side
operations over the directory tree by excluding modifications to the tree.
Also, to walk in the leaf->root direction in the tree where we don't have
a natural d_lock ordering.

This could be accomplished by taking every d_lock, but this would mean a
huge number of locks and actually gets very tricky.

Solve this instead by using the rename seqlock for multi-step read-side
operations, retry in case of a rename so we don't walk up the wrong parent.
Concurrent dentry insertions are not serialised against.  Concurrent deletes
are tricky when walking up the directory: our parent might have been deleted
when dropping locks so also need to check and retry for that.

We can also use the rename lock in cases where livelock is a worry (and it
is introduced in subsequent patch).
Signed-off-by: NNick Piggin <npiggin@kernel.dk>
      949854d0
    • N
      fs: dcache scale subdirs · 2fd6b7f5
      由 Nick Piggin 提交于 
      Protect d_subdirs and d_child with d_lock, except in filesystems that aren't
using dcache_lock for these anyway (eg. using i_mutex).

Note: if we change the locking rule in future so that ->d_child protection is
provided only with ->d_parent->d_lock, it may allow us to reduce some locking.
But it would be an exception to an otherwise regular locking scheme, so we'd
have to see some good results. Probably not worthwhile.
Signed-off-by: NNick Piggin <npiggin@kernel.dk>
      2fd6b7f5
    • N
      fs: dcache scale d_unhashed · da502956
      由 Nick Piggin 提交于 
      Protect d_unhashed(dentry) condition with d_lock. This means keeping
DCACHE_UNHASHED bit in synch with hash manipulations.
Signed-off-by: NNick Piggin <npiggin@kernel.dk>
      da502956
    • N
      fs: dcache scale dentry refcount · b7ab39f6
      由 Nick Piggin 提交于 
      Make d_count non-atomic and protect it with d_lock. This allows us to ensure a
0 refcount dentry remains 0 without dcache_lock. It is also fairly natural when
we start protecting many other dentry members with d_lock.
Signed-off-by: NNick Piggin <npiggin@kernel.dk>
      b7ab39f6
    • N
      fs: change d_hash for rcu-walk · b1e6a015
      由 Nick Piggin 提交于 
      Change d_hash so it may be called from lock-free RCU lookups. See similar
patch for d_compare for details.

For in-tree filesystems, this is just a mechanical change.
Signed-off-by: NNick Piggin <npiggin@kernel.dk>
      b1e6a015
    • N
      fs: change d_compare for rcu-walk · 621e155a
      由 Nick Piggin 提交于 
      Change d_compare so it may be called from lock-free RCU lookups. This
does put significant restrictions on what may be done from the callback,
however there don't seem to have been any problems with in-tree fses.
If some strange use case pops up that _really_ cannot cope with the
rcu-walk rules, we can just add new rcu-unaware callbacks, which would
cause name lookup to drop out of rcu-walk mode.

For in-tree filesystems, this is just a mechanical change.
Signed-off-by: NNick Piggin <npiggin@kernel.dk>
      621e155a
    • N
      fs: name case update method · fb2d5b86
      由 Nick Piggin 提交于 
      smpfs and ncpfs want to update a live dentry name in-place. Rather than
have them open code the locking, provide a documented dcache API.
Signed-off-by: NNick Piggin <npiggin@kernel.dk>
      fb2d5b86
    • N
      fs: change d_delete semantics · fe15ce44
      由 Nick Piggin 提交于 
      Change d_delete from a dentry deletion notification to a dentry caching
advise, more like ->drop_inode. Require it to be constant and idempotent,
and not take d_lock. This is how all existing filesystems use the callback
anyway.

This makes fine grained dentry locking of dput and dentry lru scanning
much simpler.
Signed-off-by: NNick Piggin <npiggin@kernel.dk>
      fe15ce44
    • R
      cassini: Use local-mac-address prom property for Cassini MAC address · 4e3dbdb1
      由 Richard Mortimer 提交于 
      Fallback on the local-mac-address prom property if the Cassini device
does not have an address programmed in the VPD ROM. This uses the same
technique as implemented by the sungem driver.

The problem was reported by Frans van Berckel using Debian kernel 2.6.34-7
on Sun Fire V440. udev was assigning a new eth<n> device name on each reboot
because the cassini driver was using a random MAC address.

Fix tested on 2.6.34-7 and 2.6.37 Sun Fire V440. Compile tested against
2.6.36 davem/sparc-2.6.git
Reported-by: NFrans van Berckel <fberckel@xs4all.nl>
Tested-by: NFrans van Berckel <fberckel@xs4all.nl>
Reviewed-by: NJulian Calaby <julian.calaby@gmail.com>
Reviewed-by: NSam Ravnborg <sam@ravnborg.org>
Signed-off-by: NRichard Mortimer <richm@oldelvet.org.uk>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      4e3dbdb1
    • D
      atyfb: Fix bootup hangs on sparc64. · 09798eb9
      由 David S. Miller 提交于 
      After commit 25edd694 ("sparc64: Get
rid of indirect p1275 PROM call buffer.")  we can't pass virtual
addresses >4GB to PROM calls.

Largely this is never necessary in drivers because we have a copy of
the entire PROM device tree in the kernel and a set of of_*()
interfaces to access it.

Unfortunately there were some lingering prom calls in the atyfb
driver, in particular prom_finddevice() was being called with an
on-stack address which could be anywhere.

This code is actually probing for information we already have, the
PROM choosen console output device is stored in of_console_device so
all of this nasty code consolidates into a one-line comparison.

Next we have some prom_getintdefault() calls which are trivially
transformed into the equivalent of_getintprop_default().

Special thanks to Fabio, who figured out exactly where the bootup
was hanging.  That made this bug trivial to fix.
Reported-by: NFabio M. Di NItto <fabbione@fabbione.net>
Reported-by: NSam Ravnborg <sam@ravnborg.org>
Reported-by: NFrans van Berckel <fberckel@xs4all.nl>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
Signed-off-by: NFabio M. Di NItto <fabbione@fabbione.net>
      09798eb9
    • F
      netconsole: clarify stopping message · 38cfb907
      由 Ferenc Wagner 提交于 
      Signed-off-by: NFerenc Wagner <wferi@niif.hu>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      38cfb907
    • F
      netconsole: don't announce stopping if nothing happened · 141dfba3
      由 Ferenc Wagner 提交于 
      Signed-off-by: NFerenc Wagner <wferi@niif.hu>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      141dfba3