Rather than keep a d_mounted count in the dentry, set a dentry flag instead.
The flag can be cleared by checking the hash table to see if there are any
mounts left, which is not time critical because it is performed at detach time.
The mounted state of a dentry is only used to speculatively take a look in the
mount hash table if it is set -- before following the mount, vfsmount lock is
taken and mount re-checked without races.
This saves 4 bytes on 32-bit, nothing on 64-bit but it does provide a hole I
might use later (and some configs have larger than 32-bit spinlocks which might
make use of the hole).
Autofs4 conversion and changelog by Ian Kent <raven@themaw.net>:
In autofs4, when expring direct (or offset) mounts we need to ensure that we
block user path walks into the autofs mount, which is covered by another mount.
To do this we clear the mounted status so that follows stop before walking into
the mount and are essentially blocked until the expire is completed. The
automount daemon still finds the correct dentry for the umount due to the
follow mount logic in fs/autofs4/root.c:autofs4_follow_link(), which is set as
an inode operation for direct and offset mounts only and is called following
the lookup that stopped at the covered mount.
At the end of the expire the covering mount probably has gone away so the
mounted status need not be restored. But we need to check this and only restore
the mounted status if the expire failed.
XXX: autofs may not work right if we have other mounts go over the top of it?
Signed-off-by: Nick Piggin <npiggin@kernel.dk>
Use a seqlock in the fs_struct to enable us to take an atomic copy of the
complete cwd and root paths. Use this in the RCU lookup path to avoid a
thread-shared spinlock in RCU lookup operations.
Multi-threaded apps may now perform path lookups with scalability matching
multi-process apps. Operations such as stat(2) become very scalable for
multi-threaded workload.
Signed-off-by: Nick Piggin <npiggin@kernel.dk>
Perform common cases of path lookups without any stores or locking in the
ancestor dentry elements. This is called rcu-walk, as opposed to the current
algorithm which is a refcount based walk, or ref-walk.
This results in far fewer atomic operations on every path element,
significantly improving path lookup performance. It also avoids cacheline
bouncing on common dentries, significantly improving scalability.
The overall design is like this:
* LOOKUP_RCU is set in nd->flags, which distinguishes rcu-walk from ref-walk.
* Take the RCU lock for the entire path walk, starting with the acquiring
of the starting path (eg. root/cwd/fd-path). So now dentry refcounts are
not required for dentry persistence.
* synchronize_rcu is called when unregistering a filesystem, so we can
access d_ops and i_ops during rcu-walk.
* Similarly take the vfsmount lock for the entire path walk. So now mnt
refcounts are not required for persistence. Also we are free to perform mount
lookups, and to assume dentry mount points and mount roots are stable up and
down the path.
* Have a per-dentry seqlock to protect the dentry name, parent, and inode,
so we can load this tuple atomically, and also check whether any of its
members have changed.
* Dentry lookups (based on parent, candidate string tuple) recheck the parent
sequence after the child is found in case anything changed in the parent
during the path walk.
* inode is also RCU protected so we can load d_inode and use the inode for
limited things.
* i_mode, i_uid, i_gid can be tested for exec permissions during path walk.
* i_op can be loaded.
When we reach the destination dentry, we lock it, recheck lookup sequence,
and increment its refcount and mountpoint refcount. RCU and vfsmount locks
are dropped. This is termed "dropping rcu-walk". If the dentry refcount does
not match, we can not drop rcu-walk gracefully at the current point in the
lokup, so instead return -ECHILD (for want of a better errno). This signals the
path walking code to re-do the entire lookup with a ref-walk.
Aside from the final dentry, there are other situations that may be encounted
where we cannot continue rcu-walk. In that case, we drop rcu-walk (ie. take
a reference on the last good dentry) and continue with a ref-walk. Again, if
we can drop rcu-walk gracefully, we return -ECHILD and do the whole lookup
using ref-walk. But it is very important that we can continue with ref-walk
for most cases, particularly to avoid the overhead of double lookups, and to
gain the scalability advantages on common path elements (like cwd and root).
The cases where rcu-walk cannot continue are:
* NULL dentry (ie. any uncached path element)
* parent with d_inode->i_op->permission or ACLs
* dentries with d_revalidate
* Following links
In future patches, permission checks and d_revalidate become rcu-walk aware. It
may be possible eventually to make following links rcu-walk aware.
Uncached path elements will always require dropping to ref-walk mode, at the
very least because i_mutex needs to be grabbed, and objects allocated.
Signed-off-by: Nick Piggin <npiggin@kernel.dk>
Add branch annotations for seqlock read fastpath, and introduce
__read_seqcount_begin and __read_seqcount_end functions, that can avoid the
smp_rmb() if used carefully. These will be used by store-free path walking
algorithm performance is critical and seqlocks are in use.
Signed-off-by: Nick Piggin <npiggin@kernel.dk>
Pseudo filesystems that don't put inode on RCU list or reachable by
rcu-walk dentries do not need to RCU free their inodes.
Signed-off-by: Nick Piggin <npiggin@kernel.dk>
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: Nick Piggin <npiggin@kernel.dk>
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: Nick Piggin <npiggin@kernel.dk>
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: Nick Piggin <npiggin@kernel.dk>
Add a new lock, dcache_inode_lock, to protect the inode's i_dentry list
from concurrent modification. d_alias is also protected by d_lock.
Signed-off-by: Nick Piggin <npiggin@kernel.dk>
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: Nick Piggin <npiggin@kernel.dk>
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: Nick Piggin <npiggin@kernel.dk>
Add a new lock, dcache_hash_lock, to protect the dcache hash table from
concurrent modification. d_hash is also protected by d_lock.
Signed-off-by: Nick Piggin <npiggin@kernel.dk>
Remove dcache_lock locking from hostfs filesystem, and move it into dcache
helpers. All that is required is a coherent path name. Protection from
concurrent modification of the namespace after path name generation is not
provided in current code, because dcache_lock is dropped before the path is
used.
Signed-off-by: Nick Piggin <npiggin@kernel.dk>
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: Nick Piggin <npiggin@kernel.dk>
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: Nick Piggin <npiggin@kernel.dk>
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: Nick Piggin <npiggin@kernel.dk>
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: Nick Piggin <npiggin@kernel.dk>
Remove redundant (and incorrect, since dcache RCU lookup) dentry locking
documentation and point to the canonical document.
Signed-off-by: Nick Piggin <npiggin@kernel.dk>
If you change the color depth via fbset or some other framebuffer aware
userland application struct fb_fix_screeninfo is not updated to this new
information. This patch fixes this issue. Also the function is changed to
just pass in struct drm_framebuffer so in the future we could use more
fields. I'm hoping some day fix->smem* could be set here :-)
Signed-off-by: James Simmons <jsimmons@infradead.org>
Signed-off-by: Dave Airlie <airlied@redhat.com>
* 'omap-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tmlind/linux-omap-2.6: (243 commits)
omap2: Make OMAP2PLUS select OMAP_DM_TIMER
OMAP4: hwmod data: Fix alignment and end of line in structurefields
OMAP4: hwmod data: Move the DMA structures
OMAP4: hwmod data: Move the smartreflex structures
OMAP4: hwmod data: Fix missing SIDLE_SMART_WKUP in smartreflexsysc
arm: omap: tusb6010: add name for MUSB IRQ
arm: omap: craneboard: Add USB EHCI support
omap2+: Initialize serial port for dynamic remuxing for n8x0
omap2+: Add struct omap_board_data and use it for platform level serial init
omap2+: Allow hwmod state changes to mux pads based on the state changes
omap2+: Add support for hwmod specific muxing of devices
omap2+: Add omap_mux_get_by_name
OMAP2: PM: fix compile error when !CONFIG_SUSPEND
MAINTAINERS: OMAP: hwmod: update hwmod code, data maintainership
OMAP4: Smartreflex framework extensions
OMAP4: hwmod: Add inital data for smartreflex modules.
OMAP4: PM: Program correct init voltages for scalable VDDs
OMAP4: Adding voltage driver support
OMAP4: Register voltage PMIC parameters with the voltage layer
OMAP3: PM: Program correct init voltages for VDD1 and VDD2
...
Fix up trivial conflict in arch/arm/plat-omap/Kconfig
* 'devel' of master.kernel.org:/home/rmk/linux-2.6-arm: (416 commits)
ARM: DMA: add support for DMA debugging
ARM: PL011: add DMA burst threshold support for ST variants
ARM: PL011: Add support for transmit DMA
ARM: PL011: Ensure IRQs are disabled in UART interrupt handler
ARM: PL011: Separate hardware FIFO size from TTY FIFO size
ARM: PL011: Allow better handling of vendor data
ARM: PL011: Ensure error flags are clear at startup
ARM: PL011: include revision number in boot-time port printk
ARM: vexpress: add sched_clock() for Versatile Express
ARM i.MX53: Make MX53 EVK bootable
ARM i.MX53: Some bug fix about MX53 MSL code
ARM: 6607/1: sa1100: Update platform device registration
ARM: 6606/1: sa1100: Fix platform device registration
ARM i.MX51: rename IPU irqs
ARM i.MX51: Add ipu clock support
ARM: imx/mx27_3ds: Add PMIC support
ARM: DMA: Replace page_to_dma()/dma_to_page() with pfn_to_dma()/dma_to_pfn()
mx51: fix usb clock support
MX51: Add support for usb host 2
arch/arm/plat-mxc/ehci.c: fix errors/typos
...
In order to enable migration support, we will want to move some of the
structures that are subject to migration into the struct nfs_server.
In particular, if we are to move the state_owner and state_owner_id to
being a per-filesystem structure, then we should label the resulting
open/lock owners with a per-filesytem label to ensure global uniqueness.
This patch does so by adding the super block s_dev to the open/lock owner
name.
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-next-2.6: (1436 commits)
cassini: Use local-mac-address prom property for Cassini MAC address
net: remove the duplicate #ifdef __KERNEL__
net: bridge: check the length of skb after nf_bridge_maybe_copy_header()
netconsole: clarify stopping message
netconsole: don't announce stopping if nothing happened
cnic: Fix the type field in SPQ messages
netfilter: fix export secctx error handling
netfilter: fix the race when initializing nf_ct_expect_hash_rnd
ipv4: IP defragmentation must be ECN aware
net: r6040: Return proper error for r6040_init_one
dcb: use after free in dcb_flushapp()
dcb: unlock on error in dcbnl_ieee_get()
net: ixp4xx_eth: Return proper error for eth_init_one
include/linux/if_ether.h: Add #define ETH_P_LINK_CTL for HPNA and wlan local tunnel
net: add POLLPRI to sock_def_readable()
af_unix: Avoid socket->sk NULL OOPS in stream connect security hooks.
net_sched: pfifo_head_drop problem
mac80211: remove stray extern
mac80211: implement off-channel TX using hw r-o-c offload
mac80211: implement hardware offload for remain-on-channel
...
Delegations are per-inode, not per-nfs_client. When a server file
system is migrated, delegations on the client must be moved from the
source to the destination nfs_server. Make it easier to manage a
mount point's delegation list across a migration event by moving the
list to the nfs_server struct.
Clean up: I added documenting comments to public functions I changed
in this patch. For consistency I added comments to all the other
public functions in fs/nfs/delegation.c.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
NFSv4 migration needs to reassociate state owners from the source to
the destination nfs_server data structures. To make that easier, move
the cl_state_owners field to the nfs_server struct. cl_openowner_id
and cl_lockowner_id accompany this move, as they are used in
conjunction with cl_state_owners.
The cl_lock field in the parent nfs_client continues to protect all
three of these fields.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
A layout can request return-on-close. How this interacts with the
forgetful model of never sending LAYOUTRETURNS is a bit ambiguous.
We forget any layouts marked roc, and wait for them to be completely
forgotten before continuing with the close. In addition, to compensate
for races with any inflight LAYOUTGETs, and the fact that we do not get
any layout stateid back from the server, we set the barrier to the worst
case scenario of current_seqid + number of outstanding LAYOUTGETS.
Signed-off-by: Fred Isaman <iisaman@netapp.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
We shouldn't send a LAYOUTGET(openstateid) unless all outstanding RPCs
using the previous stateid are completed. This requires choosing the
stateid to encode earlier, so we can abort if one is not available (we
want to use the open stateid, but a LAYOUTGET is already out using
it), and adding a count of the number of outstanding rpc calls using
layout state (which for now consist solely of LAYOUTGETs).
Signed-off-by: Fred Isaman <iisaman@netapp.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
Differentiate from server backchannel
Signed-off-by: Andy Adamson <andros@netapp.com>
Acked-by: Bruce Fields <bfields@redhat.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
Currently session draining only drains the fore channel.
The back channel processing must also be drained.
Use the back channel highest_slot_used to indicate that a callback is being
processed by the callback thread. Move the session complete to be per channel.
When the session is draininig, wait for any current back channel processing
to complete and stop all new back channel processing by returning NFS4ERR_DELAY
to the back channel client.
Drain the back channel, then the fore channel.
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
Fixes a bug where the nfs_client could be freed during callback processing.
Refactor nfs_find_client to use minorversion specific means to locate the
correct nfs_client structure.
In the NFS layer, V4.0 clients are found using the callback_ident field in the
CB_COMPOUND header. V4.1 clients are found using the sessionID in the
CB_SEQUENCE operation which is also compared against the sessionID associated
with the back channel thread after a successful CREATE_SESSION.
Each of these methods finds the one an only nfs_client associated
with the incoming callback request - so nfs_find_client_next is not needed.
In the RPC layer, the pg_authenticate call needs to find the nfs_client. For
the v4.0 callback service, the callback identifier has not been decoded so a
search by address, version, and minorversion is used. The sessionid for the
sessions based callback service has (usually) not been set for the
pg_authenticate on a CB_NULL call which can be sent prior to the return
of a CREATE_SESSION call, so the sessionid associated with the back channel
thread is not used to find the client in pg_authenticate for CB_NULL calls.
Pass the referenced nfs_client to each CB_COMPOUND operation being proceesed
via the new cb_process_state structure. The reference is held across
cb_compound processing.
Use the new cb_process_state struct to move the NFS4ERR_RETRY_UNCACHED_REP
processing from process_op into nfs4_callback_sequence where it belongs.
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
The sessions based callback service is started prior to the CREATE_SESSION call
so that it can handle CB_NULL requests which can be sent before the
CREATE_SESSION call returns and the session ID is known.
Set the callback sessionid after a sucessful CREATE_SESSION.
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
Use the small id to pointer translator service to provide a unique callback
identifier per SETCLIENTID call used to identify the v4.0 callback service
associated with the clientid.
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
Move the current sock create and destroy routines into the new transport ops.
Back channel socket will be destroyed by the svc_closs_all call in svc_destroy.
Added check: only TCP supported on shared back channel.
Signed-off-by: Andy Adamson <andros@netapp.com>
Acked-by: Bruce Fields <bfields@redhat.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
Since we are already in #ifdef __KERNEL__, we don't need to check it
again.
Signed-off-by: Changli Gao <xiaosuo@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Since nf_ct_expect_dst_hash() may be called without nf_conntrack_lock
locked, nf_ct_expect_hash_rnd should be initialized in the atomic way.
In this patch, we use nf_conntrack_hash_rnd instead of
nf_ct_expect_hash_rnd.
Signed-off-by: Changli Gao <xiaosuo@gmail.com>
Acked-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
* 'x86-alternatives-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip:
x86, suspend: Avoid unnecessary smp alternatives switch during suspend/resume
* 'x86-fpu-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip:
x86-64, asm: Use fxsaveq/fxrestorq in more places
* 'x86-hwmon-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip:
x86, hwmon: Add core threshold notification to therm_throt.c
* 'x86-paravirt-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip:
x86, paravirt: Use native_halt on a halt, not native_safe_halt
* 'core-locking-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip:
locking, lockdep: Convert sprintf_symbol to %pS
* 'irq-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip:
irq: Better struct irqaction layout
Ethertype used by HPNA control protocols (LARQ, rate, link, etc) and by
Broadcom wlan drivers for local signalling.
Signed-off-by: Henry Ptasinski <henryp@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
* 'x86-security-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip:
module: Move RO/NX module protection to after ftrace module update
x86: Resume trampoline must be executable
x86: Add RO/NX protection for loadable kernel modules
x86: Add NX protection for kernel data
x86: Fix improper large page preservation
