Make it possible for a running process (such as gssapid) to be able to
instantiate a key, as was requested by Trond Myklebust for NFS4.
The patch makes the following changes:
(1) A new, optional key type method has been added. This permits a key type
to intercept requests at the point /sbin/request-key is about to be
spawned and do something else with them - passing them over the
rpc_pipefs files or netlink sockets for instance.
The uninstantiated key, the authorisation key and the intended operation
name are passed to the method.
(2) The callout_info is no longer passed as an argument to /sbin/request-key
to prevent unauthorised viewing of this data using ps or by looking in
/proc/pid/cmdline.
This means that the old /sbin/request-key program will not work with the
patched kernel as it will expect to see an extra argument that is no
longer there.
A revised keyutils package will be made available tomorrow.
(3) The callout_info is now attached to the authorisation key. Reading this
key will retrieve the information.
(4) A new field has been added to the task_struct. This holds the
authorisation key currently active for a thread. Searches now look here
for the caller's set of keys rather than looking for an auth key in the
lowest level of the session keyring.
This permits a thread to be servicing multiple requests at once and to
switch between them. Note that this is per-thread, not per-process, and
so is usable in multithreaded programs.
The setting of this field is inherited across fork and exec.
(5) A new keyctl function (KEYCTL_ASSUME_AUTHORITY) has been added that
permits a thread to assume the authority to deal with an uninstantiated
key. Assumption is only permitted if the authorisation key associated
with the uninstantiated key is somewhere in the thread's keyrings.
This function can also clear the assumption.
(6) A new magic key specifier has been added to refer to the currently
assumed authorisation key (KEY_SPEC_REQKEY_AUTH_KEY).
(7) Instantiation will only proceed if the appropriate authorisation key is
assumed first. The assumed authorisation key is discarded if
instantiation is successful.
(8) key_validate() is moved from the file of request_key functions to the
file of permissions functions.
(9) The documentation is updated.
From: <Valdis.Kletnieks@vt.edu>
Build fix.
Signed-off-by: David Howells <dhowells@redhat.com>
Cc: Trond Myklebust <trond.myklebust@fys.uio.no>
Cc: Alexander Zangerl <az@bond.edu.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Cause any links within a keyring to keys that match a key to be linked into
that keyring to be discarded as a link to the new key is added. The match is
contingent on the type and description strings being the same.
This permits requests, adds and searches to displace negative, expired,
revoked and dead keys easily. After some discussion it was concluded that
duplicate valid keys should probably be discarded also as they would otherwise
hide the new key.
Since request_key() is intended to be the primary method by which keys are
added to a keyring, duplicate valid keys wouldn't be an issue there as that
function would return an existing match in preference to creating a new key.
Signed-off-by: David Howells <dhowells@redhat.com>
Cc: Trond Myklebust <trond.myklebust@fys.uio.no>
Cc: Alexander Zangerl <az@bond.edu.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Add a new keyctl function that allows the expiry time to be set on a key or
removed from a key, provided the caller has attribute modification access.
Signed-off-by: David Howells <dhowells@redhat.com>
Cc: Trond Myklebust <trond.myklebust@fys.uio.no>
Cc: Alexander Zangerl <az@bond.edu.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
kmsg_write returns with printk, so some programs may be confused by a
successful write() with a return value different than the buffer length.
# /bin/echo something > /dev/kmsg
/bin/echo: write error: Inappropriate ioctl for device
The drawbacks is that the printk return value can no more be quickly
checked from userspace.
Signed-off-by: Guillaume Chazarain <guichaz@yahoo.fr>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
What's the true meaning of the printk return value? Should it include the
priority prefix length of 3? and what about the timing information? In
both cases it was broken:
strace -e write echo 1 > /dev/kmsg
=> write(1, "1\n", 2) = 5
strace -e write echo "<1>1" > /dev/kmsg
=> write(1, "<1>1\n", 5) = 8
The returned length was "length of input string + 3", I made it "length
of string output to the log buffer".
Note that I couldn't find any printk caller in the kernel interested by its
return value besides kmsg_write.
Signed-off-by: Guillaume Chazarain <guichaz@yahoo.fr>
Acked-By: Tim Bird <tim.bird@am.sony.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
When making an fctl locking call through compat_sys_fcntl64 (i.e. a 32bit
app on a 64bit kernel), the syscall can return a locking range that is in
conflict with the queried lock.
If some aspect of this range does not fit in the 32bit structure, something
needs to be done.
The current code is wrong in several respects:
- It returns data to userspace even if no conflict was found
i.e. it should check l_type for F_UNLCK
- It returns -EOVERFLOW too agressively. A lock range covering
the last possible byte of the file (start = COMPAT_OFF_T_MAX,
len = 1) should be possible, but is rejected with the current test.
- A extra-long 'len' should not be a problem. If only that part
of the conflicting lock that would be visible to the 32bit
app needs to be reported to the 32bit app anyway.
This patch addresses those three issues and adds a comment to (hopefully)
record it for posterity.
Note: this patch mainly affects test-cases. Real applications rarely is
ever see the problems.
This patch has been tested (LSB test suite), and works.
Signed-off-by: Neil Brown <neilb@suse.de>
Cc: Arnd Bergmann <arnd@arndb.de>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Matthew Wilcox <willy@debian.org>
Cc: Trond Myklebust <trond.myklebust@fys.uio.no>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
SUS requires that when truncating a file to the size that it currently
is:
truncate and ftruncate should NOT modify ctime or mtime
O_TRUNC SHOULD modify ctime and mtime.
Currently mtime and ctime are always modified on most local
filesystems (side effect of ->truncate) or never modified (on NFS).
With this patch:
ATTR_CTIME|ATTR_MTIME are sent with ATTR_SIZE precisely when
an update of these times is required whether size changes or not
(via a new argument to do_truncate). This allows NFS to do
the right thing for O_TRUNC.
inode_setattr nolonger forces ATTR_MTIME|ATTR_CTIME when the ATTR_SIZE
sets the size to it's current value. This allows local filesystems
to do the right thing for f?truncate.
Also, the logic in inode_setattr is changed a bit so there are two return
points. One returns the error from vmtruncate if it failed, the other
returns 0 (there can be no other failure).
Finally, if vmtruncate succeeds, and ATTR_SIZE is the only change
requested, we now fall-through and mark_inode_dirty. If a filesystem did
not have a ->truncate function, then vmtruncate will have changed i_size,
without marking the inode as 'dirty', and I think this is wrong.
Signed-off-by: Neil Brown <neilb@suse.de>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Trond Myklebust <trond.myklebust@fys.uio.no>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Make it possible to include linux/pagevec.h multiple times without
incurring errors due to duplicate definitions.
Signed-off-by: David Howells <dhowells@redhat.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Reported from Redhat Bugzilla Bug 170450
"I updated to the development kernel and now during boot only the top of the
text is visable. For example the monitor screen the is the lines and I can
only see text in the asterisk area.
When doublescan mode is in use, scanlines must be doubled.
Thanks to Jason Dravet <dravet@hotmail.com> for reporting and testing.
Signed-off-by: Samuel Thibault <samuel.thibault@ens-lyon.org>
Cc: <stable@kernel.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
inode can never be NULL when calling this function.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
The ptrace_get_task_struct() helper that I added as part of the ptrace
consolidation is useful in variety of places that currently opencode it.
Switch them to the common helpers.
Add a ptrace_traceme() helper that needs to be explicitly called, and simplify
the ptrace_get_task_struct() interface. We don't need the request argument
now, and we return the task_struct directly, using ERR_PTR() for error
returns. It's a bit more code in the callers, but we have two sane routines
that do one thing well now.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
librelay and relay-app.h have been retired - update Documentation to reflect
that.
Signed-off-by: Tom Zanussi <zanussi@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
This patch renames relayfs_file_operations to relay_file_operations, and the
file operations themselves from relayfs_XXX to relay_file_XXX, to make it more
clear that they refer to relay files.
Signed-off-by: Tom Zanussi <zanussi@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Documentation update for creating global buffers.
Signed-off-by: Tom Zanussi <zanussi@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
This patch adds the optional is_global outparam to the create_buf_file()
callback. This can be used by clients to create a single global relayfs
buffer instead of the default per-cpu buffers. This was suggested as being
useful for certain debugging applications where it's more convenient to be
able to get all the data from a single channel without having to go to the
bother of dealing with per-cpu files.
Signed-off-by: Tom Zanussi <zanussi@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Documentation update for creating relay files in other filesystems.
Signed-off-by: Tom Zanussi <zanussi@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
This patch adds a couple of callback functions that allow a client to hook
into relay_open()/close() and supply the files that will be used to represent
the channel buffers; the default implementation if no callbacks are defined is
to create the files in relayfs. This is to support the creation and use of
relay files in other filesystems such as debugfs, as implied by the fact that
relayfs_file_operations are exported.
Signed-off-by: Tom Zanussi <zanussi@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Since we're no longer using relayfs_inode_info, remove relayfs_alloc_inode()
and relayfs_destroy_inode() along with the relayfs inode cache.
Signed-off-by: Tom Zanussi <zanussi@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Use inode->u.generic_ip instead of relayfs_inode_info to store pointer to user
data. Clients using relayfs_file_create() to create their own files would
probably more expect their data to be stored in generic_ip; we also intend in
the next set of patches to get rid of relayfs-specific stuff in the file
operations, so we might as well do it here.
Signed-off-by: Tom Zanussi <zanussi@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
This patch adds and exports relayfs_remove_file(), for API symmetry (with
relayfs_create_file()).
Signed-off-by: Tom Zanussi <zanussi@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
This patch adds a mandatory fileops param to relayfs_create_file() and exports
that function so that clients can use it to create files defined by their own
set of file operations, in relayfs. The purpose is to allow relayfs
applications to create their own set of 'control' files alongside their relay
files in relayfs rather than having to create them in /proc or debugfs for
instance. relayfs_create_file() is also used by relay_open_buf() to create
the relay files for a channel. In this case, a pointer to
relayfs_file_operations is passed in, along with a pointer to the buffer
associated with the file.
Signed-off-by: Tom Zanussi <zanussi@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
The patch series implementa or fixes 3 things that were specifically requested
or suggested by relayfs users:
- support for non-relay files (patches 1-6)
Currently, the relayfs API only supports the creation of directories
(relayfs_create_dir()) and relay files (relay_open()). These patches adds
support for non-relay files (relayfs_create_file()). This is so relayfs
applications can create 'control files' in relayfs itself rather than in /proc
or via a netlink channel, as is currently done in the relay-app examples.
Basically what this amounts to is exporting relayfs_create_file() with an
additional file_ops param that clients can use to supply file operations for
their own special-purpose files in relayfs.
- make exported relay file ops useful (patches 7-8)
The relayfs relay_file_operations have always been exported, the intent being
to make it possible to create relay files in other filesystems such as
debugfs. The problem, though, is that currently the file operations are too
tightly coupled to relayfs to actually be used for this purpose. This patch
fixes that by adding a couple of callback functions that allow a client to
hook into relay_open()/close() and supply the files that will be used to
represent the channel buffers; the default implementation if no callbacks are
defined is to create the files in relayfs.
- add an option to create global relay buffer (patches 9-10) The file creation
callback also supplies an optional param, is_global, that can be used by
clients to create a single global relayfs buffer instead of the default
per-cpu buffers. This was suggested as being useful for certain debugging
applications where it's more convenient to be able to get all the data from a
single channel without having to go to the bother of dealing with per-cpu
files.
- cleanup, some renaming and Documentation updates (patches 11-12)
There were several comments that the use of netlink in the example code was
non-intuitive and in fact the whole relay-app business was needlessly
confusing. Based on that feedback, the example code has been completely
converted over to relayfs control files as supported by this patch, and have
also been made completely self-contained.
The converted examples along with a couple of new examples that demonstrate
using exported relay files can be found in relay-apps tarball:
http://prdownloads.sourceforge.net/relayfs/relay-apps-0.9.tar.gz?download
This patch:
Separate buffer create/destroy from inode create/destroy. We want to be able
to associate other data and not just relay buffers with inodes. Buffer
create/destroy is moved out of inode.c and into relayfs core code.
Signed-off-by: Tom Zanussi <zanussi@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Needed for the Novell kernel debugger and perhaps some per-cpu data on x86_64
in the future.
Cc: Andi Kleen <ak@muc.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Use atomic_inc_not_zero for rcu files instead of special case rcuref.
Signed-off-by: Nick Piggin <npiggin@suse.de>
Cc: "Paul E. McKenney" <paulmck@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Convert atomic_dec_and_lock to use new atomic primitives.
Signed-off-by: Nick Piggin <npiggin@suse.de>
Cc: "Paul E. McKenney" <paulmck@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Use bd_claim() when opening the cdrom device to prevent user space programs
such as cdrecord, hald and kded from interfering with the burning process.
Signed-off-by: Peter Osterlund <petero2@telia.com>
Cc: Christoph Hellwig <hch@lst.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
This patch contains the following cleanups:
- make needlessly global functions static
- every file should include the headers containing the prototypes for
it's global functions
Signed-off-by: Adrian Bunk <bunk@stusta.de>
Acked-by: "Paul E. McKenney" <paulmck@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
"extern inline" -> "static inline"
Since there's no pullphone() function this patch removes the dead
prototype.
Signed-off-by: Adrian Bunk <bunk@stusta.de>
Acked-by: Karsten Keil <kkeil@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
This patch moves the rtc_interrupt() prototype to rtc.h and removes the
prototypes from C files.
It also renames static rtc_interrupt() functions in
arch/arm/mach-integrator/time.c and arch/sh64/kernel/time.c to avoid compile
problems.
Signed-off-by: Adrian Bunk <bunk@stusta.de>
Signed-off-by: Paul Gortmaker <p_gortmaker@yahoo.com>
Acked-by: Paul Mundt <lethal@linux-sh.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
This patch add EXPORT_SYMBOL(filemap_write_and_wait) and use it.
See mm/filemap.c:
And changes the filemap_write_and_wait() and filemap_write_and_wait_range().
Current filemap_write_and_wait() doesn't wait if filemap_fdatawrite()
returns error. However, even if filemap_fdatawrite() returned an
error, it may have submitted the partially data pages to the device.
(e.g. in the case of -ENOSPC)
<quotation>
Andrew Morton writes,
If filemap_fdatawrite() returns an error, this might be due to some
I/O problem: dead disk, unplugged cable, etc. Given the generally
crappy quality of the kernel's handling of such exceptions, there's a
good chance that the filemap_fdatawait() will get stuck in D state
forever.
</quotation>
So, this patch doesn't wait if filemap_fdatawrite() returns the -EIO.
Trond, could you please review the nfs part? Especially I'm not sure,
nfs must use the "filemap_fdatawrite(inode->i_mapping) == 0", or not.
Acked-by: Trond Myklebust <trond.myklebust@fys.uio.no>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
This patch changes generic_cont_expand(), in order to share the code
with fatfs.
- Use vmtruncate() if ->prepare_write() returns a error.
Even if ->prepare_write() returns an error, it may already have added some
blocks. So, this truncates blocks outside of ->i_size by vmtruncate().
- Add generic_cont_expand_simple().
The generic_cont_expand_simple() assumes that ->prepare_write() can handle
the block boundary. With this, we don't need to care the extra byte.
And for expanding a file size by truncate(), fatfs uses the
added generic_cont_expand_simple().
Signed-off-by: OGAWA Hirofumi <hirofumi@mail.parknet.co.jp>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
This exports/changes the sync_page_range/_nolock(). The fatfs needs
sync_page_range/_nolock() for expanding truncate, and changes "size_t count"
to "loff_t count".
Signed-off-by: OGAWA Hirofumi <hirofumi@mail.parknet.co.jp>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
This patch add to support of ->direct_IO() for mostly read.
The user of this seems to want to use for streaming read. So, current direct
I/O has limitation, it can only overwrite. (For write operation, mainly we
need to handle the hole etc..)
Signed-off-by: OGAWA Hirofumi <hirofumi@mail.parknet.co.jp>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
All EXPORT_SYMBOL of fatfs is only for vfat/msdos. _GPL would be proper.
Signed-off-by: OGAWA Hirofumi <hirofumi@mail.parknet.co.jp>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
We don't need to allocate buffer for checking the buffer is uptodate. This
use sb_find_get_block() instead, and if it returns NULL it's not uptodate.
Signed-off-by: OGAWA Hirofumi <hirofumi@mail.parknet.co.jp>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
It is overkill to update the FS_INFO whenever modifying
prev_free/free_clusters, because those are just a hint.
So, this patch uses ->write_super() for updating FS_INFO instead.
Signed-off-by: OGAWA Hirofumi <hirofumi@mail.parknet.co.jp>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Some ARM platforms have the ability to program the interrupt controller to
detect various interrupt edges and/or levels. For some platforms, this is
critical to setup correctly, particularly those which the setting is dependent
on the device.
Currently, ARM drivers do (eg) the following:
err = request_irq(irq, ...);
set_irq_type(irq, IRQT_RISING);
However, if the interrupt has previously been programmed to be level sensitive
(for whatever reason) then this will cause an interrupt storm.
Hence, if we combine set_irq_type() with request_irq(), we can then safely set
the type prior to unmasking the interrupt. The unfortunate problem is that in
order to support this, these flags need to be visible outside of the ARM
architecture - drivers such as smc91x need these flags and they're
cross-architecture.
Finally, the SA_TRIGGER_* flag passed to request_irq() should reflect the
property that the device would like. The IRQ controller code should do its
best to select the most appropriate supported mode.
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
New character device driver for the SyncLink GT and SyncLink AC families of
synchronous and asynchronous serial adapters
Signed-off-by: Paul Fulghum <paulkf@microgate.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Include fixes for 2.6.14-git11. Should allow to remove sched.h from
module.h on i386, x86_64, arm, ia64, ppc, ppc64, and s390. Probably more
to come since I haven't yet checked the other archs.
Signed-off-by: Tim Schmielau <tim@physik3.uni-rostock.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
For systems that aren't using cpusets, but have them CONFIG_CPUSET enabled in
their kernel (eventually this may be most distribution kernels), this patch
removes even the minimal rcu_read_lock() from the memory page allocation path.
Actually, it removes that rcu call for any task that is in the root cpuset
(top_cpuset), which on systems not actively using cpusets, is all tasks.
We don't need the rcu check for tasks in the top_cpuset, because the
top_cpuset is statically allocated, so at no risk of being freed out from
underneath us.
Signed-off-by: Paul Jackson <pj@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Mark cpuset global 'number_of_cpusets' as __read_mostly.
This global is accessed everytime a zone is considered in the zonelist loops
beneath __alloc_pages, looking for a free memory page. If number_of_cpusets
is just one, then we can short circuit the mems_allowed check.
Since this global is read alot on a hot path, and written rarely, it is an
excellent candidate for __read_mostly.
Thanks to Christoph Lameter for the suggestion.
Signed-off-by: Paul Jackson <pj@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Optimize the cpuset impact on page allocation, the most performance critical
cpuset hook in the kernel.
On each page allocation, the cpuset hook needs to check for a possible change
in the current tasks cpuset. It can now handle the common case, of no change,
without taking any spinlock or semaphore, thanks to RCU.
Convert a spinlock on the current task to an rcu_read_lock(), saving
approximately a memory barrier and an atomic op, depending on architecture.
This is done by adding rcu_assign_pointer() and synchronize_rcu() calls to the
write side of the task->cpuset pointer, in cpuset.c:attach_task(), to delay
freeing up a detached cpuset until after any critical sections referencing
that pointer.
Thanks to Andi Kleen, Nick Piggin and Eric Dumazet for ideas.
Signed-off-by: Paul Jackson <pj@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Remove a couple of more lines of code from the cpuset hooks in the page
allocation code path.
There was a check for a NULL cpuset pointer in the routine
cpuset_update_task_memory_state() that was only needed during system boot,
after the memory subsystem was initialized, before the cpuset subsystem was
initialized, to catch a NULL task->cpuset pointer.
Add a cpuset_init_early() routine, just before the mem_init() call in
init/main.c, that sets up just enough of the init tasks cpuset structure to
render cpuset_update_task_memory_state() calls harmless.
Signed-off-by: Paul Jackson <pj@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Given the mechanism in the previous patch to handle rebinding the per-vma
mempolicies of all tasks in a cpuset that changes its memory placement, it is
now easier to handle the page migration requirements of such tasks at the same
time.
The previous code didn't actually attempt to migrate the pages of the tasks in
a cpuset whose memory placement changed until the next time each such task
tried to allocate memory. This was undesirable, as users invoking memory page
migration exected to happen when the placement changed, not some unspecified
time later when the task needed more memory.
It is now trivial to handle the page migration at the same time as the per-vma
rebinding is done.
The routine cpuset.c:update_nodemask(), which handles changing a cpusets
memory placement ('mems') now checks for the special case of being asked to
write a placement that is the same as before. It was harmless enough before
to just recompute everything again, even though nothing had changed. But page
migration is a heavy weight operation - moving pages about. So now it is
worth avoiding that if asked to move a cpuset to its current location.
Signed-off-by: Paul Jackson <pj@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Fix more of longstanding bug in cpuset/mempolicy interaction.
NUMA mempolicies (mm/mempolicy.c) are constrained by the current tasks cpuset
to just the Memory Nodes allowed by that cpuset. The kernel maintains
internal state for each mempolicy, tracking what nodes are used for the
MPOL_INTERLEAVE, MPOL_BIND or MPOL_PREFERRED policies.
When a tasks cpuset memory placement changes, whether because the cpuset
changed, or because the task was attached to a different cpuset, then the
tasks mempolicies have to be rebound to the new cpuset placement, so as to
preserve the cpuset-relative numbering of the nodes in that policy.
An earlier fix handled such mempolicy rebinding for mempolicies attached to a
task.
This fix rebinds mempolicies attached to vma's (address ranges in a tasks
address space.) Due to the need to hold the task->mm->mmap_sem semaphore while
updating vma's, the rebinding of vma mempolicies has to be done when the
cpuset memory placement is changed, at which time mmap_sem can be safely
acquired. The tasks mempolicy is rebound later, when the task next attempts
to allocate memory and notices that its task->cpuset_mems_generation is
out-of-date with its cpusets mems_generation.
Because walking the tasklist to find all tasks attached to a changing cpuset
requires holding tasklist_lock, a spinlock, one cannot update the vma's of the
affected tasks while doing the tasklist scan. In general, one cannot acquire
a semaphore (which can sleep) while already holding a spinlock (such as
tasklist_lock). So a list of mm references has to be built up during the
tasklist scan, then the tasklist lock dropped, then for each mm, its mmap_sem
acquired, and the vma's in that mm rebound.
Once the tasklist lock is dropped, affected tasks may fork new tasks, before
their mm's are rebound. A kernel global 'cpuset_being_rebound' is set to
point to the cpuset being rebound (there can only be one; cpuset modifications
are done under a global 'manage_sem' semaphore), and the mpol_copy code that
is used to copy a tasks mempolicies during fork catches such forking tasks,
and ensures their children are also rebound.
When a task is moved to a different cpuset, it is easier, as there is only one
task involved. It's mm->vma's are scanned, using the same
mpol_rebind_policy() as used above.
It may happen that both the mpol_copy hook and the update done via the
tasklist scan update the same mm twice. This is ok, as the mempolicies of
each vma in an mm keep track of what mems_allowed they are relative to, and
safely no-op a second request to rebind to the same nodes.
Signed-off-by: Paul Jackson <pj@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Easy little optimization hack to avoid actually having to call
cpuset_zone_allowed() and check mems_allowed, in the main page allocation
routine, __alloc_pages(). This saves several CPU cycles per page allocation
on systems not using cpusets.
A counter is updated each time a cpuset is created or removed, and whenever
there is only one cpuset in the system, it must be the root cpuset, which
contains all CPUs and all Memory Nodes. In that case, when the counter is
one, all allocations are allowed.
Signed-off-by: Paul Jackson <pj@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
David Howells