| Age | Commit message (Collapse) | Author |
|---|
|
The mininum objects per slab is calculated based on the number of processors
that may come online.
Processors min_objects
---------------------------
1 8
2 12
4 16
8 20
16 24
32 28
64 32
1024 48
4096 56
The higher the number of processors the large the order sizes used for various
slab caches will become. This has been shown to address the performance issues
in hackbench on 16p etc.
The calculation is only performed if slub_min_objects is zero (default). If one
specifies a slub_min_objects on boot then that setting is taken.
As suggested by Zhang Yanmin's performance tests on 16-core Tigerton, use the
formula '4 * (fls(nr_cpu_ids) + 1)':
./hackbench 100 process 2000:
1) 2.6.25-rc6slab: 23.5 seconds
2) 2.6.25-rc7SLUB+slub_min_objects=20: 31 seconds
3) 2.6.25-rc7SLUB+slub_min_objects=24: 23.5 seconds
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Zhang Yanmin <yanmin.zhang@intel.com>
Signed-off-by: Pekka Enberg <penberg@cs.helsinki.fi>
|
|
We can now fallback to order 0 slabs. So set the slub_max_order to
PAGE_CACHE_ORDER_COSTLY but keep the slub_min_objects at 4. This
will mostly preserve the orders used in 2.6.25. F.e. The 2k kmalloc slab
will use order 1 allocs and the 4k kmalloc slab order 2.
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Pekka Enberg <penberg@cs.helsinki.fi>
|
|
Since we now have total_objects counter per node use that to
check for the presence of any objects. The loop over all cpu slabs
is not that useful since any cpu slab would require an object allocation
first. So drop that.
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Pekka Enberg <penberg@cs.helsinki.fi>
|
|
Makes /sys/kernel/slab/<slabname>/order writable. The allocation
order of a slab cache can then be changed dynamically during runtime.
This can be used to override the objects per slabs value establisheed
with the slub_min_objects setting that was manually specified or
calculated on bootup.
The changes of the slab order can occur while allocate_slab() runs.
Allocate slab needs the order and the number of slab objects that
are both changed by the change of order. Both are put into
a single word (struct kmem_cache_order_objects). They can then
be atomically updated and retrieved.
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Pekka Enberg <penberg@cs.helsinki.fi>
|
|
There is now a generic method of falling back to a slab page of minimal
order. No need anymore for the fallback to kmalloc_large().
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Pekka Enberg <penberg@cs.helsinki.fi>
|
|
If any higher order allocation fails then fall back the smallest order
necessary to contain at least one object. This enables fallback for all
allocations to order 0 pages. The fallback will waste more memory (objects
will not fit neatly) and the fallback slabs will be not as efficient as larger
slabs since they contain less objects.
Note that SLAB also depends on order 1 allocations for some slabs that waste
too much memory if forced into PAGE_SIZE'd page. SLUB now can now deal with
failing order 1 allocs which SLAB cannot do.
Add a new field min that will contain the objects for the smallest possible order
for a slab cache.
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Pekka Enberg <penberg@cs.helsinki.fi>
|
|
Change the statistics to consider that slabs of the same slabcache
can have different number of objects in them since they may be of
different order.
Provide a new sysfs field
total_objects
which shows the total objects that the allocated slabs of a slabcache
could hold.
Add a max field that holds the largest slab order that was ever used
for a slab cache.
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Pekka Enberg <penberg@cs.helsinki.fi>
|
|
Pack the order and the number of objects into a single word.
This saves some memory in the kmem_cache_structure and more importantly
allows us to fetch both values atomically.
Later the slab orders become runtime configurable and we need to fetch these
two items together in order to properly allocate a slab and initialize its
objects.
Fix the race by fetching the order and the number of objects in one word.
[penberg@cs.helsinki.fi: fix memset() page order in new_slab()]
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Pekka Enberg <penberg@cs.helsinki.fi>
|
|
Pass the number of objects to the for_each_object macro. Most of these are
debug related.
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Pekka Enberg <penberg@cs.helsinki.fi>
|
|
Split the inuse field up to be able to store the number of objects in this
page in the page struct as well. Necessary if we want to have pages of
various orders for a slab. Also avoids touching struct kmem_cache cachelines in
__slab_alloc().
Update diagnostic code to check the number of objects and make sure that
the number of objects always stays within the bounds of a 16 bit unsigned
integer.
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Pekka Enberg <penberg@cs.helsinki.fi>
|
|
Dump a list of unfreed objects if a slab cache is closed but
objects still remain.
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Pekka Enberg <penberg@cs.helsinki.fi>
|
|
free_list looked a bit screwy so here is an attempt to clean it up.
free_list is is only used for freeing partial lists. We do not need to return a
parameter if we decrement nr_partial within the function which allows a
simplification of the whole thing.
The current version modifies nr_partial outside of the list_lock which is
technically not correct. It was only ok because we should be the only user of
this slab cache at this point.
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Pekka Enberg <penberg@cs.helsinki.fi>
|
|
As pointed out by Ingo, the SLUB warning of calling kmem_cache_destroy()
with cache that still has objects triggers in practice. So turn this
WARN_ON() into a nice SLUB specific error message to avoid people
confusing it to a SLUB bug.
Cc: Ingo Molnar <mingo@elte.hu>
Acked-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Pekka Enberg <penberg@cs.helsinki.fi>
|
|
alignment.
This may trigger misaligned memory access exception.
Acked-by: Matt Mackall <mpm@selenic.com>
Signed-off-by: Yi Li <yi.li@analog.com>
Signed-off-by: Bryan Wu <cooloney@kernel.org>
Signed-off-by: Pekka Enberg <penberg@cs.helsinki.fi>
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/x86/linux-2.6-x86-fixes
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/x86/linux-2.6-x86-fixes:
"make namespacecheck" fixes
x86: fix compilation error in VisWS
x86: voyager fix
x86: Drop duplicate from setup.c
intel-iommu.c: dma ops fix
|
|
Signed-off-by: Ingo Molnar <mingo@elte.hu>
|
|
Signed-off-by: Alexey Starikovskiy <astarikovskiy@suse.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
|
|
Reported-by: Adrian Bunk <bunk@kernel.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
|
|
Signed-off-by: Alexey Starikovskiy <astarikovskiy@suse.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
|
|
Stephen Rothwell noticed that:
Commit 2be621498d461b63ca6124f86e3b9582e1a8e722 ("x86: dma-ops on highmem
fix") in Linus' tree introduced a new warning (noticed in the x86_64
allmodconfig build of linux-next):
drivers/pci/intel-iommu.c:2240: warning: initialization from incompatible pointer type
Which points at an instance of map_single that needs updating.
Fix it to the new prototype.
Signed-off-by: Ingo Molnar <mingo@elte.hu>
|
|
* git://git.kernel.org/pub/scm/linux/kernel/git/sfrench/cifs-2.6:
[CIFS] Fix typo in previous commit
[CIFS] Fix define for new proxy cap to match documentation
[CIFS] Fix UNC path prefix on QueryUnixPathInfo to have correct slash
[CIFS] Reserve new proxy cap for WAFS
[CIFS] Add various missing flags and defintions
[CIFS] make cifs_dfs_automount_list_static
[CIFS] Fix oops when slow oplock process races with unmount
[CIFS] Fix acl length when very short ACL being modified by chmod
[CIFS] Fix looping on reconnect to Samba when unexpected tree connect fail on reconnect
[CIFS] minor update to change log
|
|
Fix a regression in the RXKAD security module introduced in:
commit 91e916cffec7c0153c5cbaa447151862a7a9a047
Author: Al Viro <viro@ftp.linux.org.uk>
Date: Sat Mar 29 03:08:38 2008 +0000
net/rxrpc trivial annotations
A variable was declared as a 16-bit type rather than a 32-bit type.
Signed-off-by: David Howells <dhowells@redhat.com>
Acked-with-apologies-by: Al Viro <viro@ZenIV.linux.org.uk>
Signed-of-by: Linus Torvalds <torvalds@linux-foundation.org>
|
|
This snuck in through 919ee677b656c52c5f86d3d916786891220d5452
("[SPARC64]: Add NUMA support")
Signed-off-by: Sebastian Siewior <sebastian@breakpoint.cc>
Cc: David S. Miller <davem@davemloft.net>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
|
|
* git://git.linux-nfs.org/projects/trondmy/nfs-2.6: (80 commits)
SUNRPC: Invalidate the RPCSEC_GSS session if the server dropped the request
make nfs_automount_list static
NFS: remove duplicate flags assignment from nfs_validate_mount_data
NFS - fix potential NULL pointer dereference v2
SUNRPC: Don't change the RPCSEC_GSS context on a credential that is in use
SUNRPC: Fix a race in gss_refresh_upcall()
SUNRPC: Don't disconnect more than once if retransmitting NFSv4 requests
SUNRPC: Remove the unused export of xprt_force_disconnect
SUNRPC: remove XS_SENDMSG_RETRY
SUNRPC: Protect creds against early garbage collection
NFSv4: Attempt to use machine credentials in SETCLIENTID calls
NFSv4: Reintroduce machine creds
NFSv4: Don't use cred->cr_ops->cr_name in nfs4_proc_setclientid()
nfs: fix printout of multiword bitfields
nfs: return negative error value from nfs{,4}_stat_to_errno
NLM/lockd: Ensure client locking calls use correct credentials
NFS: Remove the buggy lock-if-signalled case from do_setlk()
NLM/lockd: Fix a race when cancelling a blocking lock
NLM/lockd: Ensure that nlmclnt_cancel() returns results of the CANCEL call
NLM: Remove the signal masking in nlmclnt_proc/nlmclnt_cancel
...
|
|
* 'for-linus' of git://linux-nfs.org/~bfields/linux: (52 commits)
knfsd: clear both setuid and setgid whenever a chown is done
knfsd: get rid of imode variable in nfsd_setattr
SUNRPC: Use unsigned loop and array index in svc_init_buffer()
SUNRPC: Use unsigned index when looping over arrays
SUNRPC: Update RPC server's TCP record marker decoder
SUNRPC: RPC server still uses 2.4 method for disabling TCP Nagle
NLM: don't let lockd exit on unexpected svc_recv errors (try #2)
NFS: don't let nfs_callback_svc exit on unexpected svc_recv errors (try #2)
Use a zero sized array for raw field in struct fid
nfsd: use static memory for callback program and stats
SUNRPC: remove svc_create_thread()
nfsd: fix comment
lockd: Fix stale nlmsvc_unlink_block comment
NFSD: Strip __KERNEL__ testing from unexported header files.
sunrpc: make token header values less confusing
gss_krb5: consistently use unsigned for seqnum
NFSD: Remove NFSv4 dependency on NFSv3
SUNRPC: Remove PROC_FS dependency
NFSD: Use "depends on" for PROC_FS dependency
nfsd: move most of fh_verify to separate function
...
|
|
* git://git.kernel.org/pub/scm/linux/kernel/git/mchehab/v4l-dvb: (452 commits)
V4L/DVB (7731): tuner-xc2028: fix signal strength calculus
V4L/DVB (7730): tuner-xc2028: Fix SCODE load for MTS firmwares
V4L/DVB (7729): Fix VIDIOCGAP corruption in ivtv
V4L/DVB (7728): tea5761: bugzilla #10462: tea5761 autodetection code were broken
V4L/DVB (7726): cx23885: Enable cx23417 support on the HVR1800
V4L/DVB (7725): cx23885: Add generic cx23417 hardware encoder support
V4L/DVB (7723): pvrusb2: Clean up input selection list generation in V4L interface
V4L/DVB (7722): pvrusb2: Implement FM radio support for Gotview USB2.0 DVD 2
V4L/DVB (7721): pvrusb2: Restructure cx23416 firmware loading to have a common exit point
V4L/DVB (7720): pvrusb2: Fix bad error code on cx23416 firmware load failure
V4L/DVB (7719): pvrusb2: Implement input selection enforcement
V4L/DVB (7718): pvrusb2-dvb: update Kbuild selections
V4L/DVB (7717): pvrusb2-dvb: add DVB-T support for Hauppauge pvrusb2 model 73xxx
V4L/DVB (7716): pvrusb2: clean up global functions
V4L/DVB (7715): pvrusb2: Clean out all use of __FUNCTION__
V4L/DVB (7714): pvrusb2: Fix hang on module removal
V4L/DVB (7713): pvrusb2: Implement cleaner DVB kernel thread shutdown
V4L/DVB (7712): pvrusb2: Close connect/disconnect race
V4L/DVB (7711): pvrusb2: Fix race on module unload
V4L/DVB (7710): pvrusb2: Implement critical digital streaming quirk for onair devices
...
|
|
|
|
RFC 2203 requires the server to drop the request if it believes the
RPCSEC_GSS context is out of sequence. The problem is that we have no way
on the client to know why the server dropped the request. In order to avoid
spinning forever trying to resend the request, the safe approach is
therefore to always invalidate the RPCSEC_GSS context on every major
timeout.
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
|
|
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
|
|
There are different tables for MTS firmwares. This should be taken into
account while selecting the proper firmware.
While at tuner-xc2028.h, improve some comments.
Thanks to Edward J. Sheldrake <ejs1920@yahoo.co.uk> for helping to
diagnose such troubles with PAL/I standard.
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
|
|
Frank Bennett reported that ivtv was causing skype to crash. With help
from one of their developers he showed it was a kernel problem.
VIDIOCGCAP copies a name into a fixed length buffer - ivtv uses names
that are too long and does not truncate them so corrupts a few bytes of
the app data area.
Possibly the names also want trimming but for now this should fix the
corruption case.
Signed-off-by: Alan Cox <alan@redhat.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
|
|
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
|
|
cx23885: Enable cx23417 support on the HVR1800
Signed-off-by: Steven Toth <stoth@hauppauge.com>
Signed-off-by: Michael Krufky <mkrufky@linuxtv.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
|
|
cx23885: Add generic cx23417 hardware encoder support.
Signed-off-by: Steven Toth <stoth@hauppauge.com>
Signed-off-by: Michael Krufky <mkrufky@linuxtv.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
|
|
interface
Change how list of possible pvrusb2 inputs is generated to include
only those interfaces that make sense for the interface instance.
Signed-off-by: Mike Isely <isely@pobox.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
|
|
Signed-off-by: Mike Isely <isely@pobox.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
|
|
common exit point
Signed-off-by: Mike Isely <isely@pobox.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
|
|
Signed-off-by: Mike Isely <isely@pobox.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
|
|
In the pvrusb2 driver, different interfaces (e.g. V4L, DVB) have
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
|
|
These changes are required with the addition of digital television support
for the Hauppauge HVR1900 & HVR1950, the OnAir Creator and Sasem USB HDTV
Signed-off-by: Michael Krufky <mkrufky@linuxtv.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
|
|
Signed-off-by: Michael Krufky <mkrufky@linuxtv.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
|
|
This patch contains the following cleanups:
- make the following needlessly global function static:
- pvr2_hdw_set_cur_freq()
- #if 0 the following unused global functions:
- pvr2_hdw_get_state_name()
- pvr2_hdw_get_debug_info_unlocked()
- pvr2_hdw_get_debug_info_locked()
Signed-off-by: Adrian Bunk <bunk@kernel.org>
Signed-off-by: Mike Isely <isely@pobox.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
|
|
Apparently the kernel developers no longer consider it proper
etiquette to use __FUNCTION__; everyone must instead use __func__
(even though it breaks with older compilers). And worse still, actual
effort is being expended to sweep this change throughout the kernel
source tree. Don't these people have better things to do? So...
Completely clean out all use of __FUNCTION__ from the pvrusb2 driver
(it was just in the sysfs interface). I'm not going to use __func__
either. So there.
Signed-off-by: Mike Isely <isely@pobox.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
|
|
The pvrusb2 driver was getting had by this scenario:
1. Task A calls kthread_stop() for task B.
2. Before exiting, then Task B calls kthread_stop() for task C.
The problem is, kthread_stop() wants to allocate an internal resource
to itself (i.e. acquire a lock), which won't be released until
kthread_stop() returns. But kthread_stop() won't return until task B
is dead. But task B won't die until it finishes its call to
kthread_stop() for task C, and that will block waiting on the resource
already allocated inside task A. Deadlock.
With the pvrusb2 driver, task A is the caller to pvr_exit(), task B is
the control thread run inside of pvrusb2-context.c, and task C is any
worker thread run inside of pvrusb2-hdw.c.
This problem got introduced by the previous threading setup change,
which was itself an attempt to fix a module tear-down race (which it
actually did fix). The lesson here is that a task being waited on as
part of a kthread_stop() simply cannot be allow to also issue a
kthread_stop() - or we make sure not to issue the enclosing
kthread_stop() until we know that the inner kthread_stop() has
completed first. The solution for the pvrusb2 driver is some hackish
code which changes the main control thread tear down into a two step
process. This then makes it possible to delay issuing the
kthread_stop() on the control thread until after we know that
everything has been torn down first. (And yes, we really need that
kthread_stop() because it's the only way to safely guarantee that a
module-referencing kernel thread has safely returned back out of the
module before we finally remove the module.)
Signed-off-by: Mike Isely <isely@pobox.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
|
|
Earlier fix to handle DVB feed thread aborts was overly-aggressive.
We can take better advantage of what kthread_stop() can do. This
change simplifies things.
Signed-off-by: Mike Isely <isely@pobox.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
|
|
If a disconnect happens before initialization is completed, the
pvrusb2 driver can accidentally touch dangling pointers. The whole
initialization function must be protected by the big_lock, and once
inside that lock, the initialization function should abort if it is
discovered that a disconnect has already taken place.
Signed-off-by: Mike Isely <isely@pobox.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
|
|
The pvrusb2 driver - for basically forever - was not enforcing a
proper module tear-down. Kernel threads are used inside the driver
and all must be gone before the module can be safely removed. This
changeset reimplements a chunk of pvrusb2-context.c to enforce this
correctly. Unfortunately this is not a simple fix. The new
implementation also cuts back on kernel thread usage; instead of there
being 1 control thread per instance now it's just 1 control thread
shared by all instances. (By dropping to a single thread then the
module exit function can block on its shutdown and the thread itself
can monitor and cleanly shut down all of the other instances first.)
Signed-off-by: Mike Isely <isely@pobox.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
|
|
onair devices
Implement timed measurement of encoder operation for the first time it
is run. This allows the driver to note when the encoder has been run
successfully for at least 1/4 second. On top of that implement
various bits to ensure that the encoder has been run once before
digital streaming for OnAir devices. This is done via several core
state machine tweaks.
Signed-off-by: Mike Isely <isely@pobox.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
|
|
Some tuners seem to not work in digital mode unless the encoder is
healthy. Implement a device attribute to represent this flag and
modify the core state machines to enforce this requirement.
Signed-off-by: Mike Isely <isely@pobox.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
|
|
If the device fails to stream, the feed thread will block forever
waiting for buffers. But while in this state it was not looking for
an exit condition from the driver DVB interface. This caused the
thread to jam. Implement a new stop flag (which will be set
appropriately) to tell the thread to stop.
Signed-off-by: Mike Isely <isely@pobox.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@infradead.org>
|
