Linus Torvalds [Wed, 24 Jun 2009 19:16:49 +0000 (12:16 -0700)]
Don't warn about order-1 allocations with __GFP_NOFAIL
Traditionally, we never failed small orders (even regardless of any
__GFP_NOFAIL flags), and slab will allocate order-1 allocations even for
small allocations that could fit in a single page (in order to avoid
excessive fragmentation).
Maybe we should remove this warning entirely, but before making that
judgement, at least limit it to bigger allocations.
Acked-by: Pekka Enberg <penberg@cs.helsinki.fi> Cc: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Linus Torvalds [Wed, 24 Jun 2009 17:47:38 +0000 (10:47 -0700)]
Merge branch 'upstream' of git://ftp.linux-mips.org/pub/scm/upstream-linus
* 'upstream' of git://ftp.linux-mips.org/pub/scm/upstream-linus:
Staging: octeon-ethernet: Fix race freeing transmit buffers.
Staging: octeon-ethernet: Convert to use net_device_ops.
MIPS: Cavium: Add CPU hotplugging code.
MIPS: SMP: Allow suspend and hibernation if CPU hotplug is available
MIPS: Add arch generic CPU hotplug
DMA: txx9dmac: use dma_unmap_single if DMA_COMPL_{SRC,DEST}_UNMAP_SINGLE set
MIPS: Sibyte: Fix build error if CONFIG_SERIAL_SB1250_DUART is undefined.
MIPS: MIPSsim: Fix build error if MSC01E_INT_BASE is undefined.
MIPS: Hibernation: Remove SMP TLB and cacheflushing code.
MIPS: Build fix - include <linux/smp.h> into all smp_processor_id() users.
MIPS: bug.h Build fix - include <linux/compiler.h>.
The actual in_use value was incremented twice, but thread-1 is going
to free memory based on its stale value, and will free one too many
times. The result is that memory is freed back to the kernel while
its packet is still in the transmit buffer. If the memory is
overwritten before it is transmitted, the hardware will put a valid
checksum on it and send it out (just like it does with good packets).
If by chance the TCP flags are clobbered but not the addresses or
ports, the result can be a broken TCP stream.
The fix is to track the number of freed packets in a single location
(a Fetch-and-Add Unit register). That way it can never get out of sync
with itself.
We try to free up to MAX_SKB_TO_FREE (currently 10) buffers at a time.
If fewer are available we adjust the free count with the difference.
The action of claiming buffers to free is atomic so two threads cannot
claim the same buffers.
Signed-off-by: David Daney <ddaney@caviumnetworks.com> Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
Ralf Baechle [Tue, 23 Jun 2009 09:20:56 +0000 (10:20 +0100)]
MIPS: SMP: Allow suspend and hibernation if CPU hotplug is available
The SMP implementation of suspend and hibernate depends on CPU hotplugging.
In the past we didn't have CPU hotplug so suspend and hibernation were not
possible on SMP systems.
Ralf Baechle [Tue, 23 Jun 2009 09:00:31 +0000 (10:00 +0100)]
MIPS: Add arch generic CPU hotplug
Each platform has to add support for CPU hotplugging itself by providing
suitable definitions for the cpu_disable and cpu_die of the smp_ops
methods and setting SYS_SUPPORTS_HOTPLUG_CPU. A platform should only set
SYS_SUPPORTS_HOTPLUG_CPU once all it's smp_ops definitions have the
necessary changes. This patch contains the changes to the dummy smp_ops
definition for uni-processor systems.
Ralf Baechle [Fri, 19 Jun 2009 14:01:44 +0000 (15:01 +0100)]
MIPS: Hibernation: Remove SMP TLB and cacheflushing code.
We can't perform any flushes on SMP from swsusp_arch_resume because
interrupts are disabled. A cross-CPU flush is unnecessary anyway
because all but the local CPU have already been disabled. A local
flush is not needed either because we didn't change any mappings. So
just delete the code.
Ralf Baechle [Fri, 19 Jun 2009 13:05:26 +0000 (14:05 +0100)]
MIPS: Build fix - include <linux/smp.h> into all smp_processor_id() users.
Some of the were relying into smp.h being dragged in by another header
which of course is fragile. <asm/cpu-info.h> uses smp_processor_id()
only in macros and including smp.h there leads to an include loop, so
don't change cpu-info.h.
* git://git.kernel.org/pub/scm/linux/kernel/git/agk/linux-2.6-dm: (48 commits)
dm mpath: change to be request based
dm: disable interrupt when taking map_lock
dm: do not set QUEUE_ORDERED_DRAIN if request based
dm: enable request based option
dm: prepare for request based option
dm raid1: add userspace log
dm: calculate queue limits during resume not load
dm log: fix create_log_context to use logical_block_size of log device
dm target:s introduce iterate devices fn
dm table: establish queue limits by copying table limits
dm table: replace struct io_restrictions with struct queue_limits
dm table: validate device logical_block_size
dm table: ensure targets are aligned to logical_block_size
dm ioctl: support cookies for udev
dm: sysfs add suspended attribute
dm table: improve warning message when devices not freed before destruction
dm mpath: add service time load balancer
dm mpath: add queue length load balancer
dm mpath: add start_io and nr_bytes to path selectors
dm snapshot: use barrier when writing exception store
...
Linus Torvalds [Wed, 24 Jun 2009 17:22:57 +0000 (10:22 -0700)]
Merge branch 'audit.b63' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/audit-current
* 'audit.b63' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/audit-current:
Fix rule eviction order for AUDIT_DIR
Audit: clean up all op= output to include string quoting
Audit: move audit_get_nd completely into audit_watch
audit: seperate audit inode watches into a subfile
Audit: clean up audit_receive_skb
Audit: cleanup netlink mesg handling
Audit: unify the printk of an skb when auditd not around
Audit: dereferencing krule as if it were an audit_watch
Audit: better estimation of execve record length
Audit: fix audit watch use after free
Linus Torvalds [Wed, 24 Jun 2009 17:17:07 +0000 (10:17 -0700)]
Merge branch 'release' of git://git.kernel.org/pub/scm/linux/kernel/git/lenb/linux-acpi-2.6
* 'release' of git://git.kernel.org/pub/scm/linux/kernel/git/lenb/linux-acpi-2.6: (72 commits)
asus-laptop: remove EXPERIMENTAL dependency
asus-laptop: use pr_fmt and pr_<level>
eeepc-laptop: cpufv updates
eeepc-laptop: sync eeepc-laptop with asus_acpi
asus_acpi: Deprecate in favor of asus-laptop
acpi4asus: update MAINTAINER and KConfig links
asus-laptop: platform dev as parent for led and backlight
eeepc-laptop: enable camera by default
ACPI: Rename ACPI processor device bus ID
acerhdf: Acer Aspire One fan control
ACPI: video: DMI workaround broken Acer 7720 BIOS enabling display brightness
ACPI: run ACPI device hot removal in kacpi_hotplug_wq
ACPI: Add the reference count to avoid unloading ACPI video bus twice
ACPI: DMI to disable Vista compatibility on some Sony laptops
ACPI: fix a deadlock in hotplug case
Show the physical device node of backlight class device.
ACPI: pdc init related memory leak with physical CPU hotplug
ACPI: pci_root: remove unused dev/fn information
ACPI: pci_root: simplify list traversals
ACPI: pci_root: use driver data rather than list lookup
...
Linus Torvalds [Wed, 24 Jun 2009 17:03:12 +0000 (10:03 -0700)]
Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs-2.6
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs-2.6: (23 commits)
switch xfs to generic acl caching helpers
helpers for acl caching + switch to those
switch shmem to inode->i_acl
switch reiserfs to inode->i_acl
switch reiserfs to usual conventions for caching ACLs
reiserfs: minimal fix for ACL caching
switch nilfs2 to inode->i_acl
switch btrfs to inode->i_acl
switch jffs2 to inode->i_acl
switch jfs to inode->i_acl
switch ext4 to inode->i_acl
switch ext3 to inode->i_acl
switch ext2 to inode->i_acl
add caching of ACLs in struct inode
fs: Add new pre-allocation ioctls to vfs for compatibility with legacy xfs ioctls
cleanup __writeback_single_inode
... and the same for vfsmount id/mount group id
Make allocation of anon devices cheaper
update Documentation/filesystems/Locking
devpts: remove module-related code
...
* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-next-2.6:
bnx2: Fix the behavior of ethtool when ONBOOT=no
qla3xxx: Don't sleep while holding lock.
qla3xxx: Give the PHY time to come out of reset.
ipv4 routing: Ensure that route cache entries are usable and reclaimable with caching is off
net: Move rx skb_orphan call to where needed
ipv6: Use correct data types for ICMPv6 type and code
net: let KS8842 driver depend on HAS_IOMEM
can: let SJA1000 driver depend on HAS_IOMEM
netxen: fix firmware init handshake
netxen: fix build with without CONFIG_PM
netfilter: xt_rateest: fix comparison with self
netfilter: xt_quota: fix incomplete initialization
netfilter: nf_log: fix direct userspace memory access in proc handler
netfilter: fix some sparse endianess warnings
netfilter: nf_conntrack: fix conntrack lookup race
netfilter: nf_conntrack: fix confirmation race condition
netfilter: nf_conntrack: death_by_timeout() fix
Linus Torvalds [Wed, 24 Jun 2009 16:57:10 +0000 (09:57 -0700)]
Merge branch 'for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jack/linux-udf-2.6
* 'for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jack/linux-udf-2.6:
udf: remove redundant tests on unsigned
udf: Use device size when drive reported bogus number of written blocks
Jiri Slaby [Wed, 24 Jun 2009 17:35:05 +0000 (18:35 +0100)]
tty: fix tty_port_block_til_ready waiting
Since commit 3e3b5c087799e536871c8261b05bc28e4783c8da ("tty: use
prepare/finish_wait"), tty_port_block_til_ready() is using
prepare_to_wait()/finish_wait(). Those functions require that the
wait_queue_t be initialised with .func=autoremove_wake_function, via
DEFINE_WAIT().
But the conversion from DECLARE_WAITQUEUE() to DEFINE_WAIT() was not made,
so this code will oops in finish_wait().
Signed-off-by: Jiri Slaby <jirislaby@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Alan Cox <alan@linux.intel.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Arjan van de Ven [Wed, 24 Jun 2009 17:34:58 +0000 (18:34 +0100)]
pci: use pci_ioremap_bar() in drivers/serial
Use the newly introduced pci_ioremap_bar() function in drivers/serial.
pci_ioremap_bar() just takes a pci device and a bar number, with the goal
of making it really hard to get wrong, while also having a central place
to stick sanity checks.
Signed-off-by: Arjan van de Ven <arjan@linux.intel.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Alan Cox <alan@linux.intel.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Atsushi Nemoto [Wed, 24 Jun 2009 17:34:34 +0000 (18:34 +0100)]
kgdb: kgdboc console poll hooks for serial_txx9 uart
Implement the serial polling hooks for the serial_txx9 uart for use with
kgdboc.
This patch once got SOB from Jason on Jul 2008 and (perhaps) merged into
kgdb-next branch, but lost somewhere then. I resend it now with Jason's
Acked-by.
Signed-off-by: Atsushi Nemoto <anemo@mba.ocn.ne.jp> Acked-by: Jason Wessel <jason.wessel@windriver.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Alan Cox <alan@linux.intel.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Al Viro [Tue, 9 Jun 2009 00:46:54 +0000 (20:46 -0400)]
reiserfs: minimal fix for ACL caching
reiserfs uses NULL as "unknown" and ERR_PTR(-ENODATA) as "no ACL";
several codepaths store the former instead of the latter.
All those codepaths go through iset_acl() and all cases when it's
called with NULL acl are for the second variety, so the minimal
fix is to teach iset_acl() to deal with that.
Proper fix is to switch to more usual conventions and avoid back
and forth between internally used ERR_PTR(-ENODATA) and NULL
expected by the rest of the kernel.
Ankit Jain [Fri, 19 Jun 2009 18:28:07 +0000 (14:28 -0400)]
fs: Add new pre-allocation ioctls to vfs for compatibility with legacy xfs ioctls
This patch adds ioctls to vfs for compatibility with legacy XFS
pre-allocation ioctls (XFS_IOC_*RESVP*). The implementation
effectively invokes sys_fallocate for the new ioctls.
Also handles the compat_ioctl case.
Note: These legacy ioctls are also implemented by OCFS2.
[AV: folded fixes from hch]
Signed-off-by: Ankit Jain <me@ankitjain.org> Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
There is no reason to for the split between __writeback_single_inode and
__sync_single_inode, the former just does a couple of checks before
tail-calling the latter. So merge the two, and while we're at it split
out the I_SYNC waiting case for data integrity writers, as it's
logically separate function. Finally rename __writeback_single_inode to
writeback_single_inode.
Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Al Viro [Wed, 24 Jun 2009 06:05:18 +0000 (02:05 -0400)]
Make allocation of anon devices cheaper
Standard trick - add a new variable (start) such that
for each n < start n is known to be busy. Allocation can
skip checking everything in [0..start) and if it returns
n, we can set start to n + 1. Freeing below start sets
start to what we'd just freed.
Of course, it still sucks if we do something like
free 0
allocate
allocate
in a loop - still O(n^2) time. However, on saner loads it
improves the things a lot and the entire thing is not worth
the trouble of switching to something with better worst-case
behaviour.
The rules for locking in many superblock operations has changed
significantly, so update the documentation for it. Also correct some
older updates and ommissions.
Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
H. Peter Anvin [Wed, 17 Jun 2009 04:15:04 +0000 (21:15 -0700)]
devpts: remove module-related code
These days, the devpts filesystem is closely integrated with the pty
memory management, and cannot be built as a module, even less removed
from the kernel. Accordingly, remove all module-related stuff from
this filesystem.
[ v2: only remove code that's actually dead ]
Signed-off-by: H. Peter Anvin <hpa@zytor.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
reiserfs: remove stray unlock_super in reiserfs_resize
Reiserfs doesn't use lock_super anywhere internally, and ->remount_fs
which calls reiserfs_resize does have it currently but also expects it
to be held on return, so there's no business for the unlock_super here.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Acked by Edward Shishkin <edward.shishkin@gmail.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Ooiwa Naohiro <nooiwa@miraclelinux.com> Acked-by: Michael Chan <mchan@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Corentin Chary [Tue, 16 Jun 2009 19:28:56 +0000 (19:28 +0000)]
eeepc-laptop: cpufv updates
Limit cpufv input to acceptables values.
Add an available_cpufv file to show available
presets.
Change cpufv ouput format from %d to %#x, it won't
break compatibility with existing userspace tools, but
it provide a more human readable output.
Signed-off-by: Corentin Chary <corentincj@iksaif.net> Signed-off-by: Len Brown <len.brown@intel.com>
Corentin Chary [Tue, 16 Jun 2009 19:28:52 +0000 (19:28 +0000)]
eeepc-laptop: sync eeepc-laptop with asus_acpi
In the default Eee PC distribution, there is a modified
asus_acpi driver. eeepc-laptop is a cleaned version of this
driver. Sync ASL enum and getter/setters with asus_acpi.
Signed-off-by: Corentin Chary <corentincj@iksaif.net> Signed-off-by: Len Brown <len.brown@intel.com>
Corentin Chary [Tue, 16 Jun 2009 19:28:47 +0000 (19:28 +0000)]
acpi4asus: update MAINTAINER and KConfig links
The bug tracker have moved from sourceforge to
http://dev.iksaif.net . The homepage of the project
is now http://acpi4asus.sf.net with links to the new
bug tracker. No change for the mailing list.
Signed-off-by: Corentin Chary <corentincj@iksaif.net> Signed-off-by: Len Brown <len.brown@intel.com>
Pekka Enberg [Tue, 16 Jun 2009 19:28:45 +0000 (19:28 +0000)]
eeepc-laptop: enable camera by default
If we leave the camera disabled by default, userspace programs (e.g.
Skype, Cheese) leave the user out in the cold saying that the machine
"has no camera." Therefore, it's better to enable camera by default and
let people who really don't want it just disable the thing.
To reduce power usage you should enable USB autosuspend:
echo -n auto > /sys/bus/usb/drivers/uvcvideo/*:*/../power/level
Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Pekka Enberg <penberg@cs.helsinki.fi> Signed-off-by: Corentin Chary <corentincj@iksaif.net> Signed-off-by: Len Brown <len.brown@intel.com>
Rename the processor device bus id. And the new bus id will be
generated as the following format:
CPU+ CPU ID
For example: If the cpu ID is 5, then the bus ID will be "CPU5".
If the CPU ID is 10, then the bus ID will be "CPUA".
Yes, this will change the directory names seen
in /proc/acpi/processor/* on some systems.
Before this patch, those directory names where
totally arbitrary strings based on the interal AML device strings.
http://bugzilla.kernel.org/show_bug.cgi?id=13612
Signed-off-by: Zhao Yakui <yakui.zhao@intel.com> Signed-off-by: Len Brown <len.brown@intel.com>
Peter Feuerer [Sun, 21 Jun 2009 16:53:03 +0000 (18:53 +0200)]
acerhdf: Acer Aspire One fan control
Acerhdf is a driver for Acer Aspire One netbooks. It allows
to access the temperature sensor and to control the fan.
Signed-off-by: Peter Feuerer <peter@piie.net> Signed-off-by: Andreas Mohr <andi@lisas.de> Signed-off-by: Borislav Petkov <petkovbb@gmail.com> Signed-off-by: Len Brown <len.brown@intel.com>
Al Viro [Wed, 24 Jun 2009 04:02:38 +0000 (00:02 -0400)]
Fix rule eviction order for AUDIT_DIR
If syscall removes the root of subtree being watched, we
definitely do not want the rules refering that subtree
to be destroyed without the syscall in question having
a chance to match them.
Eric Paris [Thu, 11 Jun 2009 18:31:37 +0000 (14:31 -0400)]
Audit: clean up all op= output to include string quoting
A number of places in the audit system we send an op= followed by a string
that includes spaces. Somehow this works but it's just wrong. This patch
moves all of those that I could find to be quoted.
Eric Paris [Thu, 11 Jun 2009 18:31:36 +0000 (14:31 -0400)]
Audit: move audit_get_nd completely into audit_watch
audit_get_nd() is only used by audit_watch and could be more cleanly
implemented by having the audit watch functions call it when needed rather
than making the generic audit rule parsing code deal with those objects.
Eric Paris [Thu, 11 Jun 2009 18:31:36 +0000 (14:31 -0400)]
audit: seperate audit inode watches into a subfile
In preparation for converting audit to use fsnotify instead of inotify we
seperate the inode watching code into it's own file. This is similar to
how the audit tree watching code is already seperated into audit_tree.c
Eric Paris [Thu, 11 Jun 2009 18:31:35 +0000 (14:31 -0400)]
Audit: clean up audit_receive_skb
audit_receive_skb is hard to clearly parse what it is doing to the netlink
message. Clean the function up so it is easy and clear to see what is going
on.
Eric Paris [Thu, 11 Jun 2009 18:31:34 +0000 (14:31 -0400)]
Audit: dereferencing krule as if it were an audit_watch
audit_update_watch() runs all of the rules for a given watch and duplicates
them, attaches a new watch to them, and then when it finishes that process
and has called free on all of the old rules (ok maybe still inside the rcu
grace period) it proceeds to use the last element from list_for_each_entry_safe()
as if it were a krule rather than being the audit_watch which was anchoring
the list to output a message about audit rules changing.
This patch unfies the audit message from two different places into a helper
function and calls it from the correct location in audit_update_rules(). We
will now get an audit message about the config changing for each rule (with
each rules filterkey) rather than the previous garbage.
Eric Paris [Thu, 11 Jun 2009 18:31:34 +0000 (14:31 -0400)]
Audit: better estimation of execve record length
The audit execve record splitting code estimates the length of the message
generated. But it forgot to include the "" that wrap each string in its
estimation. This means that execve messages with lots of tiny (1-2 byte)
arguments could still cause records greater than 8k to be emitted. Simply
fix the estimate.
Eric Paris [Thu, 11 Jun 2009 18:31:33 +0000 (14:31 -0400)]
Audit: fix audit watch use after free
When an audit watch is added to a parent the temporary watch inside the
original krule from userspace is freed. Yet the original watch is used after
the real watch was created in audit_add_rules()
Zhao Yakui [Tue, 16 Jun 2009 03:23:13 +0000 (11:23 +0800)]
ACPI: Add the reference count to avoid unloading ACPI video bus twice
Sometimes both acpi video and i915 driver are compiled as modules.
And there exists the strict dependency between the two drivers.
The acpi video bus will be unloaded in course of unloading the i915 driver.
If we unload the acpi video driver, then the kernel oops will be triggered.
Add the reference count to avoid unloading the ACPI video bus twice.
The reference count should be checked before unregistering the acpi video bus.
If the reference count is already zero, it won't unregister it again.
And after the acpi video bus is already unregistered, the reference count
will be set to zero.
http://bugzilla.kernel.org/show_bug.cgi?id=13396
Signed-off-by: Zhao Yakui <yakui.zhao@intel.com> Acked-by: Zhang Rui <rui.zhang@intel.com> Signed-off-by: Len Brown <len.brown@intel.com>
Zhang Rui [Mon, 22 Jun 2009 03:31:17 +0000 (11:31 +0800)]
ACPI: DMI to disable Vista compatibility on some Sony laptops
Linux claims Vista compatibility to the BIOS for a number of
reasons, but this brings hard lockup on some Sony laptops.
Disable Vista compatibility via DMI for these laptops unless
we can figure out what Vista is doing for this platform.
http://bugzilla.kernel.org/show_bug.cgi?id=12904
Signed-off-by: Zhang Rui <rui.zhang@intel.com> Signed-off-by: Len Brown <len.brown@intel.com>
Zhang Rui [Tue, 23 Jun 2009 02:20:29 +0000 (10:20 +0800)]
ACPI: fix a deadlock in hotplug case
we used to run the hotplug code in keventd_wq.
But when hot removing the ACPI battery device,
power_supply_unregister invokes flush_scheduled_work.
This causes a deadlock. i.e
1. When dock is unplugged, all the hotplug code is run on kevent_wq.
2. the hotplug code removes all the child devices of dock device.
3. removing the child device may invoke flush_scheduled_work
4. flush_scheduled_work waits until all the work on kevent_wq to be
finished, while this will never be true because the hotplug code
is running on keventd_wq...
Introduce a new workqueue for hotplug in this patch.
http://bugzilla.kernel.org/show_bug.cgi?id=13533
Tested-by: Paul Martin <pm@debian.org> Tested-by: Vojtech Gondzala <vojtech.gondzala@gmail.com> Signed-off-by: Zhang Rui <rui.zhang@intel.com> Reviewed-by: Bjorn Helgaas <bjorn.helgaas@hp.com> Signed-off-by: Len Brown <len.brown@intel.com>
Linus Torvalds [Wed, 24 Jun 2009 02:49:24 +0000 (19:49 -0700)]
Merge git://git.infradead.org/iommu-2.6
* git://git.infradead.org/iommu-2.6:
Intel-IOMMU, intr-remap: source-id checking
Intel-IOMMU, intr-remap: set the whole 128bits of irte when modify/free it
IOMMU Identity Mapping Support (drivers/pci/intel_iommu.c)
Linus Torvalds [Wed, 24 Jun 2009 02:36:02 +0000 (19:36 -0700)]
Merge branch 'upstream-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jlbec/ocfs2
* 'upstream-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jlbec/ocfs2:
ocfs2/trivial: Wrap ocfs2_sysfile_cluster_lock_key within define.
ocfs2: Add lockdep annotations
vfs: Set special lockdep map for dirs only if not set by fs
ocfs2: Disable orphan scanning for local and hard-ro mounts
ocfs2: Do not initialize lvb in ocfs2_orphan_scan_lock_res_init()
ocfs2: Stop orphan scan as early as possible during umount
ocfs2: Fix ocfs2_osb_dump()
ocfs2: Pin journal head before accessing jh->b_committed_data
ocfs2: Update atime in splice read if necessary.
ocfs2: Provide the ocfs2_dlm_lvb_valid() stack API.
Neil Horman [Mon, 22 Jun 2009 10:18:53 +0000 (10:18 +0000)]
ipv4 routing: Ensure that route cache entries are usable and reclaimable with caching is off
When route caching is disabled (rt_caching returns false), We still use route
cache entries that are created and passed into rt_intern_hash once. These
routes need to be made usable for the one call path that holds a reference to
them, and they need to be reclaimed when they're finished with their use. To be
made usable, they need to be associated with a neighbor table entry (which they
currently are not), otherwise iproute_finish2 just discards the packet, since we
don't know which L2 peer to send the packet to. To do this binding, we need to
follow the path a bit higher up in rt_intern_hash, which calls
arp_bind_neighbour, but not assign the route entry to the hash table.
Currently, if caching is off, we simply assign the route to the rp pointer and
are reutrn success. This patch associates us with a neighbor entry first.
Secondly, we need to make sure that any single use routes like this are known to
the garbage collector when caching is off. If caching is off, and we try to
hash in a route, it will leak when its refcount reaches zero. To avoid this,
this patch calls rt_free on the route cache entry passed into rt_intern_hash.
This places us on the gc list for the route cache garbage collector, so that
when its refcount reaches zero, it will be reclaimed (Thanks to Alexey for this
suggestion).
I've tested this on a local system here, and with these patches in place, I'm
able to maintain routed connectivity to remote systems, even if I set
/proc/sys/net/ipv4/rt_cache_rebuild_count to -1, which forces rt_caching to
return false.
Signed-off-by: Neil Horman <nhorman@redhat.com> Reported-by: Jarek Poplawski <jarkao2@gmail.com> Reported-by: Maxime Bizon <mbizon@freebox.fr> Signed-off-by: David S. Miller <davem@davemloft.net>
Herbert Xu [Mon, 22 Jun 2009 02:25:25 +0000 (02:25 +0000)]
net: Move rx skb_orphan call to where needed
In order to get the tun driver to account packets, we need to be
able to receive packets with destructors set. To be on the safe
side, I added an skb_orphan call for all protocols by default since
some of them (IP in particular) cannot handle receiving packets
destructors properly.
Now it seems that at least one protocol (CAN) expects to be able
to pass skb->sk through the rx path without getting clobbered.
So this patch attempts to fix this properly by moving the skb_orphan
call to where it's actually needed. In particular, I've added it
to skb_set_owner_[rw] which is what most users of skb->destructor
call.
This is actually an improvement for tun too since it means that
we only give back the amount charged to the socket when the skb
is passed to another socket that will also be charged accordingly.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au> Tested-by: Oliver Hartkopp <olver@hartkopp.net> Signed-off-by: David S. Miller <davem@davemloft.net>
Baruch Siach [Mon, 22 Jun 2009 13:36:29 +0000 (16:36 +0300)]
i2c: driver for the Synopsys DesignWare I2C controller
The i2c Linux driver for the DesignWare i2c block of Synopsys, which is meant
for AMBA Peripheral Bus. This i2c block is used on SoC chips like the ARM9
based PVG610.
Signed-off-by: Baruch Siach <baruch@tkos.co.il> Signed-off-by: Ben Dooks <ben-linux@fluff.org>
Weidong Han [Fri, 22 May 2009 16:41:15 +0000 (00:41 +0800)]
Intel-IOMMU, intr-remap: source-id checking
To support domain-isolation usages, the platform hardware must be
capable of uniquely identifying the requestor (source-id) for each
interrupt message. Without source-id checking for interrupt remapping
, a rouge guest/VM with assigned devices can launch interrupt attacks
to bring down anothe guest/VM or the VMM itself.
This patch adds source-id checking for interrupt remapping, and then
really isolates interrupts for guests/VMs with assigned devices.
Because PCI subsystem is not initialized yet when set up IOAPIC
entries, use read_pci_config_byte to access PCI config space directly.
Signed-off-by: Weidong Han <weidong.han@intel.com> Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
Weidong Han [Fri, 22 May 2009 16:41:14 +0000 (00:41 +0800)]
Intel-IOMMU, intr-remap: set the whole 128bits of irte when modify/free it
Interrupt remapping table entry is 128bits. Currently, it only sets low
64bits of irte in modify_irte and free_irte. This ignores high 64bits
setting of irte, that means source-id setting will be ignored. This patch
sets the whole 128bits of irte when modify/free it. Following source-id
checking patch depends on this.
Signed-off-by: Weidong Han <weidong.han@intel.com> Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
Fenghua Yu [Fri, 19 Jun 2009 20:47:29 +0000 (13:47 -0700)]
IOMMU Identity Mapping Support (drivers/pci/intel_iommu.c)
Identity mapping for IOMMU defines a single domain to 1:1 map all PCI
devices to all usable memory.
This reduces map/unmap overhead in DMA API's and improve IOMMU
performance. On 10Gb network cards, Netperf shows no performance
degradation compared to non-IOMMU performance.
This method may lose some of DMA remapping benefits like isolation.
The patch sets up identity mapping for all PCI devices to all usable
memory. In the DMA API, there is no overhead to maintain page tables,
invalidate iotlb, flush cache etc.
32 bit DMA devices don't use identity mapping domain, in order to access
memory beyond 4GiB.
When kernel option iommu=pt, pass through is first tried. If pass
through succeeds, IOMMU goes to pass through. If pass through is not
supported in hw or fail for whatever reason, IOMMU goes to identity
mapping.
Signed-off-by: Fenghua Yu <fenghua.yu@intel.com> Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
Robert P. J. Day [Tue, 23 Jun 2009 19:37:06 +0000 (12:37 -0700)]
h8/300: fix incorrect "select" directives in arch/h8300/Kconfig.cpu.
Fix the incorrect "select" directives by dropping the "CONFIG_"
prefixes, and correcting the typo "H8S2768" to "H8S2678".
Signed-off-by: Robert P. J. Day <rpjday@crashcourse.ca> Acked-by: Yoshinori Sato <ysato@users.sourceforge.jp> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Dimitri Sivanich [Tue, 23 Jun 2009 19:37:04 +0000 (12:37 -0700)]
mm: fix handling of pagesets for downed cpus
After downing/upping a cpu, an attempt to set
/proc/sys/vm/percpu_pagelist_fraction results in an oops in
percpu_pagelist_fraction_sysctl_handler().
If a processor is downed then we need to set the pageset pointer back to
the boot pageset.
Updates of the high water marks should not access pagesets of unpopulated
zones (those pointer go to the boot pagesets which would be no longer
functional if their size would be increased beyond zero).
Signed-off-by: Dimitri Sivanich <sivanich@sgi.com> Signed-off-by: Christoph Lameter <cl@linux-foundation.org> Reviewed-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Cc: Nick Piggin <nickpiggin@yahoo.com.au> Cc: Mel Gorman <mel@csn.ul.ie> Cc: <stable@kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Mike Frysinger [Tue, 23 Jun 2009 19:37:01 +0000 (12:37 -0700)]
rmap: fixup page_referenced() for nommu systems
After the recent changes that went into mm/vmscan.c to overhaul stuff, we
ended up with these warnings on no-mmu systems:
mm/vmscan.c: In function `shrink_page_list':
mm/vmscan.c:580: warning: unused variable `vm_flags'
mm/vmscan.c: In function `shrink_active_list':
mm/vmscan.c:1294: warning: `vm_flags' may be used uninitialized in this function
mm/vmscan.c:1242: note: `vm_flags' was declared here
This is because the no-mmu function defines page_referenced() to work on
the first argument only (the page). It does not clear the vm_flags given
to it because for no-mmu systems, they never actually get utilized. Since
that is no longer strictly true, we need to set vm_flags to 0 like
everyone else so gcc can do proper dead code elimination without annoying
us with unused warnings.
Signed-off-by: Mike Frysinger <vapier@gentoo.org> Cc: David Howells <dhowells@redhat.com> Acked-by: David McCullough <davidm@snapgear.com> Cc: Greg Ungerer <gerg@uclinux.org> Cc: Paul Mundt <lethal@linux-sh.org> Cc: Wu Fengguang <fengguang.wu@intel.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Hugh Dickins [Tue, 23 Jun 2009 19:36:58 +0000 (12:36 -0700)]
mm: pass mm to grab_swap_token
If a kthread happens to use get_user_pages() on an mm (as KSM does),
there's a chance that it will end up trying to read in a swap page, then
oops in grab_swap_token() because the kthread has no mm: GUP passes down
the right mm, so grab_swap_token() ought to be using it.
We have not identified a stronger case than KSM's daemon (not yet in
mainline), but the issue must have come up before, since RHEL has included
a fix for this for years (though a different fix, they just back out of
grab_swap_token if current->mm is unset: which is what we first proposed,
but using the right mm here seems more correct).
Reported-by: Izik Eidus <ieidus@redhat.com> Signed-off-by: Johannes Weiner <hannes@cmpxchg.org> Signed-off-by: Hugh Dickins <hugh.dickins@tiscali.co.uk> Acked-by: Rik van Riel <riel@redhat.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* git://git.kernel.org/pub/scm/linux/kernel/git/lethal/sh-2.6:
sh: Fix up HAVE_PERF_COUNTERS typo.
sh: Fix up more dma-mapping fallout.
sh: SH7786 SMP support.
sh: Wire up the uncached fixmap on sh64 as well.
sh: Use local TLB flush in set_pte_phys().
sh: Provide cpu_idle_wait() to fix up cpuidle/SMP build.
Linus Torvalds [Tue, 23 Jun 2009 18:36:01 +0000 (11:36 -0700)]
Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/roland/infiniband
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/roland/infiniband:
RDMA: Add __init/__exit macros to addr.c and cma.c
IB/ehca: Bump version number
mlx4_core: Fix dma_sync_single_for_cpu() with matching for_device() calls
IB/mthca: Replace dma_sync_single() use with proper functions
RDMA/nes: Fix FIN state handling under error conditions
RDMA/nes: Fix max_qp_init_rd_atom returned from query device
IB/ehca: Ensure that guid_entry index is not negative
IB/ehca: Tolerate dynamic memory operations before driver load
* git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi-misc-2.6: (38 commits)
fusion: mptsas, fix lock imbalance
[SCSI] scsi_transport_fc: replace BUS_ID_SIZE by fixed count
sd, sr: fix Driver 'sd' needs updating message
scsi_transport_iscsi: return -EOVERFLOW for Too many iscsi targets
fc_transport: Selective return value from BSG timeout function
fc_transport: The softirq_done function registration for BSG request
sym53c8xx: ratelimit parity errors
explain the hidden scsi_wait_scan Kconfig variable
ibmvfc: Fix endless PRLI loop in discovery
ibmvfc: Process async events before command responses
libfc: Add runtime debugging with debug_logging module parameter
libfcoe: Add runtime debugging with module param debug_logging
fcoe: Add runtime debug logging with module parameter debug_logging
scsi_debug: Add support for physical block exponent and alignment
cnic: add NETDEV_1000 and NETDEVICES to Kconfig select
cnic: Fix __symbol_get() build error.
Revert "[SCSI] cnic: fix error: implicit declaration of function ‘__symbol_get’"
ipr: differentiate pci-x and pci-e based adapters
ipr: add test for MSI interrupt support
scsi_transport_spi: Blacklist Ultrium-3 tape for IU transfers
...
Linus Torvalds [Tue, 23 Jun 2009 18:34:24 +0000 (11:34 -0700)]
Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/arnd/asm-generic
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/arnd/asm-generic:
asm-generic: add dummy pgprot_noncached()
lib/checksum.c: fix endianess bug
asm-generic: hook up new system calls
asm-generic: list Arnd as asm-generic maintainer
asm-generic: drop HARDIRQ_BITS definition from hardirq.h
asm-generic: uaccess: fix up local access_ok() usage
asm-generic: uaccess: add missing access_ok() check to strnlen_user()