openwrt/staging/blogic.git
14 years agovirtio: console: Add ability to remove module
Amit Shah [Fri, 12 Feb 2010 05:02:16 +0000 (10:32 +0530)]
virtio: console: Add ability to remove module

Add the ability to remove the virtio_console module.

This aids debugging.

Signed-off-by: Amit Shah <amit.shah@redhat.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
14 years agovirtio: console: Ensure no memleaks in case of unused buffers
Amit Shah [Fri, 12 Feb 2010 05:02:15 +0000 (10:32 +0530)]
virtio: console: Ensure no memleaks in case of unused buffers

If unused data exists in in_vq, ensure we flush that first and then
detach unused buffers, which will ensure all buffers from the in_vq are
removed.

Also ensure we free the buffers after detaching them.

Signed-off-by: Amit Shah <amit.shah@redhat.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
14 years agovirtio: console: show error message if hvc_alloc fails for console ports
Amit Shah [Mon, 18 Jan 2010 11:05:23 +0000 (16:35 +0530)]
virtio: console: show error message if hvc_alloc fails for console ports

Signed-off-by: Amit Shah <amit.shah@redhat.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
14 years agovirtio: console: Add debugfs files for each port to expose debug info
Amit Shah [Mon, 21 Dec 2009 17:06:21 +0000 (22:36 +0530)]
virtio: console: Add debugfs files for each port to expose debug info

This is helpful in examining ports' state.

Signed-off-by: Amit Shah <amit.shah@redhat.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
14 years agovirtio: console: Add ability to hot-unplug ports
Amit Shah [Mon, 21 Dec 2009 16:57:31 +0000 (22:27 +0530)]
virtio: console: Add ability to hot-unplug ports

Remove port data; deregister from the hvc core if it's a console port.

Signed-off-by: Amit Shah <amit.shah@redhat.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
14 years agovirtio: console: Handle port hot-plug
Amit Shah [Mon, 21 Dec 2009 16:52:08 +0000 (22:22 +0530)]
virtio: console: Handle port hot-plug

If the 'nr_ports' variable in the config space is updated to a higher
value, that means new ports have been hotplugged.

Introduce a new workqueue to handle such updates and create new ports.

Signed-off-by: Amit Shah <amit.shah@redhat.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
14 years agovirtio: console: Remove cached data on port close
Amit Shah [Mon, 21 Dec 2009 16:45:30 +0000 (22:15 +0530)]
virtio: console: Remove cached data on port close

Remove any data that we might have in a port's inbuf when closing a port
or when any data is received when a port is closed.

Signed-off-by: Amit Shah <amit.shah@redhat.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
14 years agovirtio: console: Register with sysfs and create a 'name' attribute for ports
Amit Shah [Mon, 21 Dec 2009 16:27:40 +0000 (21:57 +0530)]
virtio: console: Register with sysfs and create a 'name' attribute for ports

The host can set a name for ports so that they're easily discoverable
instead of going by the /dev/vportNpn naming. This attribute will be
placed in /sys/class/virtio-ports/vportNpn/name. udev scripts can then
create symlinks to the port using the name.

Signed-off-by: Amit Shah <amit.shah@redhat.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
14 years agovirtio: console: Ensure only one process can have a port open at a time
Amit Shah [Thu, 26 Nov 2009 05:55:38 +0000 (11:25 +0530)]
virtio: console: Ensure only one process can have a port open at a time

Add a guest_connected field that ensures only one process
can have a port open at a time.

This also ensures we don't have a race when we later add support for
dropping buffers when closing the char dev and buffer caching is turned
off for the particular port.

Signed-off-by: Amit Shah <amit.shah@redhat.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
14 years agovirtio: console: Add file operations to ports for open/read/write/poll
Amit Shah [Mon, 21 Dec 2009 16:19:30 +0000 (21:49 +0530)]
virtio: console: Add file operations to ports for open/read/write/poll

Allow guest userspace applications to open, read from, write to, poll
the ports via the char dev interface.

When a port gets opened, a notification is sent to the host via a
control message indicating a connection has been established. Similarly,
on closing of the port, a notification is sent indicating disconnection.

Signed-off-by: Amit Shah <amit.shah@redhat.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
14 years agovirtio: console: Associate each port with a char device
Amit Shah [Mon, 21 Dec 2009 16:06:04 +0000 (21:36 +0530)]
virtio: console: Associate each port with a char device

The char device will be used as an interface by applications on the
guest to communicate with apps on the host.

The devices created are placed in /dev/vportNpn where N is the
virtio-console device number and n is the port number for that device.

One dynamic major device number is allocated for each device and minor
numbers are allocated for the ports contained within that device.

The file operation for the char devs will be added in the following
commits.

Signed-off-by: Amit Shah <amit.shah@redhat.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
14 years agovirtio: console: Prepare for writing to userspace buffers
Amit Shah [Mon, 21 Dec 2009 15:56:45 +0000 (21:26 +0530)]
virtio: console: Prepare for writing to userspace buffers

When ports get advertised as char devices, the buffers will come from
userspace. Equip the fill_readbuf function with the ability to write
to userspace buffers.

Signed-off-by: Amit Shah <amit.shah@redhat.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
14 years agovirtio: console: Add a new MULTIPORT feature, support for generic ports
Amit Shah [Mon, 21 Dec 2009 15:33:25 +0000 (21:03 +0530)]
virtio: console: Add a new MULTIPORT feature, support for generic ports

This commit adds a new feature, MULTIPORT. If the host supports this
feature as well, the config space has the number of ports defined for
that device. New ports are spawned according to this information.

The config space also has the maximum number of ports that can be
spawned for a particular device. This is useful in initializing the
appropriate number of virtqueues in advance, as ports might be
hot-plugged in later.

Using this feature, generic ports can be created which are not tied to
hvc consoles.

We also open up a private channel between the host and the guest via
which some "control" messages are exchanged for the ports, like whether
the port being spawned is a console port, resizing the console window,
etc.

Next commits will add support for hotplugging and presenting char
devices in /dev/ for bi-directional guest-host communication.

Signed-off-by: Amit Shah <amit.shah@redhat.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
14 years agovirtio: console: Introduce a send_buf function for a common path for sending data...
Amit Shah [Mon, 21 Dec 2009 11:58:51 +0000 (17:28 +0530)]
virtio: console: Introduce a send_buf function for a common path for sending data to host

Adding support for generic ports that will write to userspace will need
some code changes.

Consolidate the write routine into send_buf() and put_chars() now just
calls into the new function.

Signed-off-by: Amit Shah <amit.shah@redhat.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
14 years agovirtio: console: Introduce function to hand off data from host to readers
Amit Shah [Mon, 18 Jan 2010 13:45:12 +0000 (19:15 +0530)]
virtio: console: Introduce function to hand off data from host to readers

In preparation for serving data to userspace (generic ports) as well as
in-kernel users (hvc consoles), separate out the functionality common to
both in a 'fill_readbuf()' function.

Signed-off-by: Amit Shah <amit.shah@redhat.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
14 years agovirtio: console: Separate out find_vqs operation into a different function
Amit Shah [Mon, 18 Jan 2010 13:45:11 +0000 (19:15 +0530)]
virtio: console: Separate out find_vqs operation into a different function

With support for multiple ports, each port will have its own input and
output vqs. Prepare the probe function for this change.

Signed-off-by: Amit Shah <amit.shah@redhat.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
14 years agovirtio: console: Separate out console init into a new function
Amit Shah [Mon, 18 Jan 2010 13:45:10 +0000 (19:15 +0530)]
virtio: console: Separate out console init into a new function

Console ports could be hot-added. Also, with the new multiport support,
a port is identified as a console port only if the host sends a control
message.

Move the console port init into a separate function so it can be invoked
from other places.

Signed-off-by: Amit Shah <amit.shah@redhat.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
14 years agovirtio: console: Separate out console-specific data into a separate struct
Amit Shah [Mon, 18 Jan 2010 13:45:09 +0000 (19:15 +0530)]
virtio: console: Separate out console-specific data into a separate struct

Move out console-specific stuff into a separate struct from 'struct
port' as we need to maintain two lists: one for all the ports (which
includes consoles) and one only for consoles since the hvc callbacks
only give us the vtermno.

This makes console handling cleaner.

Signed-off-by: Amit Shah <amit.shah@redhat.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
14 years agovirtio: console: ensure console size is updated on hvc open
Amit Shah [Mon, 18 Jan 2010 13:45:08 +0000 (19:15 +0530)]
virtio: console: ensure console size is updated on hvc open

When multiple console support is added, ensure each port's size gets
updated when a new one is opened via hvc.

Signed-off-by: Amit Shah <amit.shah@redhat.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
14 years agovirtio: console: struct ports for multiple ports per device.
Amit Shah [Mon, 18 Jan 2010 13:45:07 +0000 (19:15 +0530)]
virtio: console: struct ports for multiple ports per device.

Rather than assume a single port, add a 'struct ports_device' which
stores data related to all the ports for that device.

Currently, there's only one port and is hooked up with hvc, but that
will change.

Signed-off-by: Amit Shah <amit.shah@redhat.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
14 years agovirtio: console: remove global var
Rusty Russell [Mon, 18 Jan 2010 13:45:06 +0000 (19:15 +0530)]
virtio: console: remove global var

Now we can use an allocation function to remove our global console variable.

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Amit Shah <amit.shah@redhat.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
14 years agovirtio: console: don't assume a single console port.
Amit Shah [Mon, 18 Jan 2010 13:45:05 +0000 (19:15 +0530)]
virtio: console: don't assume a single console port.

Keep a list of all ports being used as a console, and provide a lock
and a lookup function.  The hvc callbacks only give us a vterm number,
so we need to map this.

Signed-off-by: Amit Shah <amit.shah@redhat.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
14 years agovirtio: console: use vdev->priv to avoid accessing global var.
Rusty Russell [Mon, 18 Jan 2010 13:45:04 +0000 (19:15 +0530)]
virtio: console: use vdev->priv to avoid accessing global var.

Part of removing our "one console" assumptions, use vdev->priv to point
to the port (currently == the global console).

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Amit Shah <amit.shah@redhat.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
14 years agovirtio: console: introduce a get_inbuf helper to fetch bufs from in_vq
Amit Shah [Mon, 18 Jan 2010 13:45:03 +0000 (19:15 +0530)]
virtio: console: introduce a get_inbuf helper to fetch bufs from in_vq

This makes taking locks around the get_buf vq operation easier, as well
as complements the add_inbuf() operation.

Signed-off-by: Amit Shah <amit.shah@redhat.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
14 years agovirtio: console: ensure add_inbuf can work for multiple ports as well
Amit Shah [Mon, 18 Jan 2010 13:45:02 +0000 (19:15 +0530)]
virtio: console: ensure add_inbuf can work for multiple ports as well

add_inbuf() assumed one port and one inbuf per port. Remove that
assumption.

Also move the function so that put_chars and get_chars are together.

Signed-off-by: Amit Shah <amit.shah@redhat.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
14 years agovirtio: console: encapsulate buffer information in a struct
Amit Shah [Mon, 18 Jan 2010 13:45:01 +0000 (19:15 +0530)]
virtio: console: encapsulate buffer information in a struct

Collect port buffer, used_len, offset fields into a single structure.

Signed-off-by: Amit Shah <amit.shah@redhat.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
14 years agovirtio: console: port encapsulation
Rusty Russell [Mon, 18 Jan 2010 13:45:00 +0000 (19:15 +0530)]
virtio: console: port encapsulation

We are heading towards a multiple-"port" system, so as part of weaning off
globals we encapsulate the information into 'struct port'.

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Amit Shah <amit.shah@redhat.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
14 years agovirtio: console: We support only one device at a time
Amit Shah [Mon, 18 Jan 2010 13:44:59 +0000 (19:14 +0530)]
virtio: console: We support only one device at a time

We support only one virtio_console device at a time. If multiple are
found, error out if one is already initialized.

Signed-off-by: Amit Shah <amit.shah@redhat.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
14 years agohvc_console: make the ops pointer const.
Rusty Russell [Sat, 28 Nov 2009 06:50:26 +0000 (12:20 +0530)]
hvc_console: make the ops pointer const.

This is nicer for modern R/O protection.  And noone needs it non-const, so
constify the callers as well.

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Amit Shah <amit.shah@redhat.com>
To: Christian Borntraeger <borntraeger@de.ibm.com>
Cc: linuxppc-dev@ozlabs.org
14 years agovirtio: console: statically initialize virtio_cons
Rusty Russell [Mon, 18 Jan 2010 13:44:56 +0000 (19:14 +0530)]
virtio: console: statically initialize virtio_cons

That way, we can make it const as is good kernel style.  We use a separate
indirection for the early console, rather than mugging ops.put_chars.

We rename it hv_ops, too.

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Amit Shah <amit.shah@redhat.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
14 years agovirtio: console: comment cleanup
Rusty Russell [Mon, 18 Jan 2010 13:44:55 +0000 (19:14 +0530)]
virtio: console: comment cleanup

Remove old lguest-style comments.

[Amit: - wingify comments acc. to kernel style
       - indent comments ]

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Amit Shah <amit.shah@redhat.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
14 years agovirtio: Initialize vq->data entries to NULL
Amit Shah [Fri, 12 Feb 2010 05:02:14 +0000 (10:32 +0530)]
virtio: Initialize vq->data entries to NULL

vq operations depend on vq->data[i] being NULL to figure out if the vq
entry is in use (since the previous patch).

We have to initialize them to NULL to ensure we don't work with junk
data and trigger false BUG_ONs.

Signed-off-by: Amit Shah <amit.shah@redhat.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Cc: Shirley Ma <xma@us.ibm.com>
14 years agovirtio: Add ability to detach unused buffers from vrings
Shirley Ma [Mon, 18 Jan 2010 13:45:23 +0000 (19:15 +0530)]
virtio: Add ability to detach unused buffers from vrings

There's currently no way for a virtio driver to ask for unused
buffers, so it has to keep a list itself to reclaim them at shutdown.
This is redundant, since virtio_ring stores that information.  So
add a new hook to do this.

Signed-off-by: Shirley Ma <xma@us.ibm.com>
Signed-off-by: Amit Shah <amit.shah@redhat.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
14 years agovirtio_blk: add block topology support
Christoph Hellwig [Wed, 24 Feb 2010 20:22:25 +0000 (14:22 -0600)]
virtio_blk: add block topology support

Allow reading various alignment values from the config page.  This
allows the guest to much better align I/O requests depending on the
storage topology.

Note that the formats for the config values appear a bit messed up,
but we follow the formats used by ATA and SCSI so they are expected in
the storage world.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
14 years agovirtio: use smp_XX barriers on SMP
Michael S. Tsirkin [Wed, 27 Jan 2010 22:42:23 +0000 (00:42 +0200)]
virtio: use smp_XX barriers on SMP

virtio is communicating with a virtual "device" that actually runs on
another host processor. Thus SMP barriers can be used to control
memory access ordering.

Where possible, we should use SMP barriers which are more lightweight than
mandatory barriers, because mandatory barriers also control MMIO effects on
accesses through relaxed memory I/O windows (which virtio does not use)
(compare specifically smp_rmb and rmb on x86_64).

We can't just use smp_mb and friends though, because
we must force memory ordering even if guest is UP since host could be
running on another CPU, but SMP barriers are defined to barrier() in
that configuration. So, for UP fall back to mandatory barriers instead.

Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
14 years agovirtio: remove bogus barriers from DEBUG version of virtio_ring.c
Rusty Russell [Wed, 24 Feb 2010 20:22:22 +0000 (14:22 -0600)]
virtio: remove bogus barriers from DEBUG version of virtio_ring.c

With DEBUG defined, we add an ->in_use flag to detect if the caller
invokes two virtio methods in parallel.  The barriers attempt to ensure
timely update of the ->in_use flag.

But they're voodoo: if we need these barriers it implies that the
calling code doesn't have sufficient synchronization to ensure the
code paths aren't invoked at the same time anyway, and we want to
detect it.

Also, adding barriers changes timing, so turning on debug has more
chance of hiding real problems.

Thanks to MST for drawing my attention to this code...

CC: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
14 years agolguest: remove unneeded zlib.h include in example launcher
Rusty Russell [Wed, 24 Feb 2010 20:22:18 +0000 (14:22 -0600)]
lguest: remove unneeded zlib.h include in example launcher

Two years ago 5bbf89fc2608 removed the horrible bzImage unpacking code.
Now it's time to remove the unneeded zlib.h include, too.

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
14 years agovirtio: fix balloon without VIRTIO_BALLOON_F_STATS_VQ
Rusty Russell [Wed, 24 Feb 2010 20:22:14 +0000 (14:22 -0600)]
virtio: fix balloon without VIRTIO_BALLOON_F_STATS_VQ

When running under qemu-kvm-0.11.0:

BUG: unable to handle kernel paging request at 56e58955
...
Process vballoon (pid: 1297, ti=c7976000 task=c70a6ca0 task.ti=c7
...
Call Trace:
 [<c88253a3>] ? balloon+0x1b3/0x440 [virtio_balloon]
 [<c041c2d7>] ? schedule+0x327/0x9d0
 [<c88251f0>] ? balloon+0x0/0x440 [virtio_balloon]
 [<c014a2d4>] ? kthread+0x74/0x80
 [<c014a260>] ? kthread+0x0/0x80
 [<c0103b36>] ? kernel_thread_helper+0x6/0x30

need_stats_update should be zero-initialized.

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Acked-by: Adam Litke <agl@us.ibm.com>
14 years agovirtio: Fix scheduling while atomic in virtio_balloon stats
Adam Litke [Thu, 10 Dec 2009 22:35:15 +0000 (16:35 -0600)]
virtio: Fix scheduling while atomic in virtio_balloon stats

This is a fix for my earlier patch: "virtio: Add memory statistics reporting to
the balloon driver (V4)".

I discovered that all_vm_events() can sleep and therefore stats collection
cannot be done in interrupt context.  One solution is to handle the interrupt
by noting that stats need to be collected and waking the existing vballoon
kthread which will complete the work via stats_handle_request().  Rusty, is
this a saner way of doing business?

There is one issue that I would like a broader opinion on.  In stats_request, I
update vb->need_stats_update and then wake up the kthread.  The kthread uses
vb->need_stats_update as a condition variable.  Do I need a memory barrier
between the update and wake_up to ensure that my kthread sees the correct
value?  My testing suggests that it is not needed but I would like some
confirmation from the experts.

Signed-off-by: Adam Litke <agl@us.ibm.com>
To: Rusty Russell <rusty@rustcorp.com.au>
Cc: Anthony Liguori <aliguori@linux.vnet.ibm.com>
Cc: linux-kernel@vger.kernel.org
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
14 years agovirtio: Add memory statistics reporting to the balloon driver (V4)
Adam Litke [Mon, 30 Nov 2009 16:14:15 +0000 (10:14 -0600)]
virtio: Add memory statistics reporting to the balloon driver (V4)

Changes since V3:
 - Do not do endian conversions as they will be done in the host
 - Report stats that reference a quantity of memory in bytes
 - Minor coding style updates

Changes since V2:
 - Increase stat field size to 64 bits
 - Report all sizes in kb (not pages)
 - Drop anon_pages stat and fix endianness conversion

Changes since V1:
 - Use a virtqueue instead of the device config space

When using ballooning to manage overcommitted memory on a host, a system for
guests to communicate their memory usage to the host can provide information
that will minimize the impact of ballooning on the guests.  The current method
employs a daemon running in each guest that communicates memory statistics to a
host daemon at a specified time interval.  The host daemon aggregates this
information and inflates and/or deflates balloons according to the level of
host memory pressure.  This approach is effective but overly complex since a
daemon must be installed inside each guest and coordinated to communicate with
the host.  A simpler approach is to collect memory statistics in the virtio
balloon driver and communicate them directly to the hypervisor.

This patch enables the guest-side support by adding stats collection and
reporting to the virtio balloon driver.

Signed-off-by: Adam Litke <agl@us.ibm.com>
Cc: Anthony Liguori <anthony@codemonkey.ws>
Cc: virtualization@lists.linux-foundation.org
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au> (minor fixes)
14 years agoAdd __devexit_p around reference to virtio_pci_remove
Jamie Lokier [Fri, 8 Jan 2010 22:01:43 +0000 (22:01 +0000)]
Add __devexit_p around reference to virtio_pci_remove

This is needed to compile with CONFIG_VIRTIO_PCI=y,
because virtio_pci_remove is marked __devexit.

Signed-off-by: Jamie Lokier <jamie@shareable.org>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
14 years agoMerge branch 'upstream' of git://ftp.linux-mips.org/pub/scm/upstream-linus
Linus Torvalds [Tue, 23 Feb 2010 03:51:39 +0000 (19:51 -0800)]
Merge branch 'upstream' of git://ftp.linux-mips.org/upstream-linus

* 'upstream' of git://ftp.linux-mips.org/pub/scm/upstream-linus:
  MIPS: BCM47xx: Fix 128MB RAM support
  MIPS: Highmem: Fix build error

14 years agoMerge branch 'parisc/tracehook' of git://git.kernel.org/pub/scm/linux/kernel/git...
Linus Torvalds [Tue, 23 Feb 2010 03:51:13 +0000 (19:51 -0800)]
Merge branch 'parisc/tracehook' of git://git./linux/kernel/git/frob/linux-2.6-roland

* 'parisc/tracehook' of git://git.kernel.org/pub/scm/linux/kernel/git/frob/linux-2.6-roland:
  Revert "parisc: HAVE_ARCH_TRACEHOOK"

14 years agofs/exec.c: fix initial stack reservation
Michael Neuling [Mon, 22 Feb 2010 20:44:24 +0000 (12:44 -0800)]
fs/exec.c: fix initial stack reservation

803bf5ec259941936262d10ecc84511b76a20921 ("fs/exec.c: restrict initial
stack space expansion to rlimit") attempts to limit the initial stack to
20*PAGE_SIZE.  Unfortunately, in attempting ensure the stack is not
reduced in size, we ended up not changing the stack at all.

This size reduction check is not necessary as the expand_stack call does
this already.

This caused a regression in UML resulting in most guest processes being
killed.

Signed-off-by: Michael Neuling <mikey@neuling.org>
Reviewed-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Acked-by: WANG Cong <xiyou.wangcong@gmail.com>
Cc: Anton Blanchard <anton@samba.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: James Morris <jmorris@namei.org>
Cc: Serge Hallyn <serue@us.ibm.com>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Jouni Malinen <j@w1.fi>
Cc: <stable@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
14 years agoefifb: fix framebuffer handoff
Marcin Slusarz [Mon, 22 Feb 2010 20:44:22 +0000 (12:44 -0800)]
efifb: fix framebuffer handoff

Commit 4410f3910947dcea8672280b3adecd53cec4e85e ("fbdev: add support for
handoff from firmware to hw framebuffers") didn't add fb_destroy
operation to efifb.  Fix it and change aperture_size to match size
passed to request_mem_region.

Addresses http://bugzilla.kernel.org/show_bug.cgi?id=15151

Signed-off-by: Marcin Slusarz <marcin.slusarz@gmail.com>
Reported-by: Alex Zhavnerchik <alex.vizor@gmail.com>
Tested-by: Alex Zhavnerchik <alex.vizor@gmail.com>
Acked-by: Peter Jones <pjones@redhat.com>
Cc: Huang Ying <ying.huang@intel.com>
Cc: Dave Airlie <airlied@redhat.com>
Cc: "Rafael J. Wysocki" <rjw@sisk.pl>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
14 years agogeode-mfgpt: restore previous behavior for selecting IRQ
Jens Rottmann [Mon, 22 Feb 2010 20:44:20 +0000 (12:44 -0800)]
geode-mfgpt: restore previous behavior for selecting IRQ

geode-mfgpt: restore previous behavior for selecting IRQ

The MFGPT IRQ used to be, in order of decreasing priority,
 * IRQ supplied by the user as a boot-time parameter,
 * IRQ previously set by the BIOS or another driver,
 * default IRQ given at compile time.

Return to this behavior, which got broken when splitting the
MFGPT/clocksource driver for 2.6.33-rc1.

Signed-off-by: Jens Rottmann <JRottmann@LiPPERTEmbedded.de>
Acked-by: Andres Salomon <dilinger@collabora.co.uk>
Cc: Jordan Crouse <jordan.crouse@amd.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: john stultz <johnstul@us.ibm.com>
Acked-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
14 years agoidr: fix a critical misallocation bug, take#2
Tejun Heo [Mon, 22 Feb 2010 20:44:19 +0000 (12:44 -0800)]
idr: fix a critical misallocation bug, take#2

This is retry of reverted 859ddf09743a8cc680af33f7259ccd0fd36bfe9d
("idr: fix a critical misallocation bug") which contained two bugs.

* pa[idp->layers] should be cleared even if it's not used by
  sub_alloc() because it's used by mark idr_mark_full().

* The original condition check also assigned pa[l] to p which the new
  code didn't do thus leaving p pointing at the wrong layer.

Both problems have been fixed and the idr code has received good amount
testing using userland testing setup where simple bitmap allocator is
run parallel to verify the result of idr allocation.

The bug this patch fixes is caused by sub_alloc() optimization path
bypassing out-of-room condition check and restarting allocation loop
with starting value higher than maximum allowed value.  For detailed
description, please read commit message of 859ddf09.

Signed-off-by: Tejun Heo <tj@kernel.org>
Based-on-patch-from: Eric Paris <eparis@redhat.com>
Reported-by: Eric Paris <eparis@redhat.com>
Tested-by: Stefan Lippers-Hollmann <s.l-h@gmx.de>
Tested-by: Serge Hallyn <serue@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
14 years agokernel/sys.c: fix missing rcu protection for sys_getpriority()
Tetsuo Handa [Mon, 22 Feb 2010 20:44:16 +0000 (12:44 -0800)]
kernel/sys.c: fix missing rcu protection for sys_getpriority()

find_task_by_vpid() is not safe without rcu_read_lock().  2.6.33-rc7 got
RCU protection for sys_setpriority() but missed it for sys_getpriority().

Signed-off-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: "Paul E. McKenney" <paulmck@us.ibm.com>
Acked-by: Serge Hallyn <serue@us.ibm.com>
Acked-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
14 years agomemcg: fix oom killing a child process in an other cgroup
KAMEZAWA Hiroyuki [Mon, 22 Feb 2010 20:44:14 +0000 (12:44 -0800)]
memcg: fix oom killing a child process in an other cgroup

Presently the oom-killer is memcg aware and it finds the worst process
from processes under memcg(s) in oom.  Then, it kills victim's child
first.

It may kill a child in another cgroup and may not be any help for
recovery.  And it will break the assumption users have.

This patch fixes it.

Signed-off-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Reviewed-by: Minchan Kim <minchan.kim@gmail.com>
Cc: Balbir Singh <balbir@linux.vnet.ibm.com>
Reviewed-by: Daisuke Nishimura <nishimura@mxp.nes.nec.co.jp>
Acked-by: David Rientjes <rientjes@google.com>
Cc: <stable@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
14 years agoMIPS: BCM47xx: Fix 128MB RAM support
Hauke Mehrtens [Sat, 20 Feb 2010 18:51:20 +0000 (19:51 +0100)]
MIPS: BCM47xx: Fix 128MB RAM support

Ignoring the last page when ddr size is 128M. Cached accesses to last page
is causing the processor to prefetch using address above 128M stepping out
of the DDR address space.

Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Cc: linux-mips@linux-mips.org
Patchwork: http://patchwork.linux-mips.org/patch/981/
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
14 years agoMIPS: Highmem: Fix build error
Yoichi Yuasa [Sat, 20 Feb 2010 12:23:22 +0000 (21:23 +0900)]
MIPS: Highmem: Fix build error

arch/mips/mm/highmem.c: In function 'kmap_init':
arch/mips/mm/highmem.c:130: error: 'init_mm' undeclared (first use in this function)
arch/mips/mm/highmem.c:130: error: (Each undeclared identifier is reported only once
arch/mips/mm/highmem.c:130: error: for each function it appears in.)

Signed-off-by: Yoichi Yuasa <yuasa@linux-mips.org>
Cc: linux-mips <linux-mips@linux-mips.org>
Patchwork: http://patchwork.linux-mips.org/patch/980/
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
14 years agoRevert "parisc: HAVE_ARCH_TRACEHOOK"
Roland McGrath [Mon, 22 Feb 2010 18:43:43 +0000 (10:43 -0800)]
Revert "parisc: HAVE_ARCH_TRACEHOOK"

This reverts commit 81bf550d9cdfe0325eb1504b06c9f6511b442c1a.

HAVE_ARCH_TRACEHOOK requires defining the user_regset interfaces,
including task_user_regset_view().  parisc doesn't do that yet,
so don't lie about it.

Signed-off-by: Roland McGrath <roland@redhat.com>
14 years agoMerge branch 'perf-fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Mon, 22 Feb 2010 16:55:32 +0000 (08:55 -0800)]
Merge branch 'perf-fixes-for-linus' of git://git./linux/kernel/git/tip/linux-2.6-tip

* 'perf-fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip:
  perf probe: Init struct probe_point and set counter correctly
  hw-breakpoint: Keep track of dr7 local enable bits
  hw-breakpoints: Accept breakpoints on NULL address
  perf_events: Fix FORK events

14 years agoMerge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/dtor/input
Linus Torvalds [Mon, 22 Feb 2010 16:48:06 +0000 (08:48 -0800)]
Merge branch 'for-linus' of git://git./linux/kernel/git/dtor/input

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/dtor/input:
  Input: usbtouchscreen - extend coordinate range for Generaltouch devices
  Input: polldev can cause crash in case when polling disabled

14 years agoi915 / PM: Fix hibernate regression caused by suspend/resume splitting
Rafael J. Wysocki [Thu, 18 Feb 2010 22:06:27 +0000 (23:06 +0100)]
i915 / PM: Fix hibernate regression caused by suspend/resume splitting

Commit 84b79f8d2882b0a84330c04839ed4d3cefd2ff77 (drm/i915: Fix crash
while aborting hibernation) attempted to fix a regression introduced
by commit cbda12d77ea590082edb6d30bd342a67ebc459e0 (drm/i915:
implement new pm ops for i915), but it went too far trying to split
the freeze/suspend and resume/thaw parts of the code.  As a result,
it introduced another regression, which only is visible on some systems.

Fix the problem by merging i915_drm_suspend() with
i915_drm_freeze() and moving some code from i915_resume()
into i915_drm_thaw(), so that intel_opregion_free() and
intel_opregion_init() are also executed in the freeze and thaw code
paths, respectively.

Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
Reported-and-tested-by: Pedro Ribeiro <pedrib@gmail.com>
Tested-by: Tino Keitel <tino.keitel@tikei.de>
Acked-by: Eric Anholt <eric@anholt.net>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
14 years agoperf probe: Init struct probe_point and set counter correctly
Masami Hiramatsu [Thu, 18 Feb 2010 18:16:52 +0000 (13:16 -0500)]
perf probe: Init struct probe_point and set counter correctly

Clear struct probe_point before using it in
show_perf_probe_events(), and set pp->found counter correctly in
synthesize_perf_probe_point(). Without this initialization,
clear_probe_point() will free random addresses.

Signed-off-by: Masami Hiramatsu <mhiramat@redhat.com>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: systemtap <systemtap@sources.redhat.com>
Cc: DLE <dle-develop@lists.sourceforge.net>
LKML-Reference: <20100218181652.26547.57790.stgit@dhcp-100-2-132.bos.redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
14 years agoInput: usbtouchscreen - extend coordinate range for Generaltouch devices
Roy Yin [Mon, 22 Feb 2010 06:52:49 +0000 (22:52 -0800)]
Input: usbtouchscreen - extend coordinate range for Generaltouch devices

Generaltouch protocol allows for coordinates in [0, 0xffff] range and
there are devices reporting coordinates as high as 0x7fff so let's update
the driver to reflect that.

Signed-off-by: Roy Yin <yhch@generaltouch.com>
Signed-off-by: Dmitry Torokhov <dtor@mail.ru>
14 years agoMerge git://git.kernel.org/pub/scm/linux/kernel/git/davem/sparc-2.6
Linus Torvalds [Mon, 22 Feb 2010 03:45:24 +0000 (19:45 -0800)]
Merge git://git./linux/kernel/git/davem/sparc-2.6

* git://git.kernel.org/pub/scm/linux/kernel/git/davem/sparc-2.6:
  sparc32: Fix struct stat uid/gid types.

14 years agosparc32: Fix struct stat uid/gid types.
David S. Miller [Mon, 22 Feb 2010 02:03:16 +0000 (18:03 -0800)]
sparc32: Fix struct stat uid/gid types.

Commit 085219f79cad89291699bd2bfb21c9fdabafe65f
("sparc32: use proper types in struct stat")

Accidently changed the struct stat uid/gid members
to uid_t and gid_t, but those get set to
__kernel_uid32_t and __kernel_gid32_t respectively.
Those are of type 'int' but the structure is meant
to have 'short'.  So use uid16_t and gid16_t to
correct this.

Reported-by: Rob Landley <rob@landley.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agoMerge git://git.kernel.org/pub/scm/linux/kernel/git/wim/linux-2.6-watchdog
Linus Torvalds [Sun, 21 Feb 2010 19:25:47 +0000 (11:25 -0800)]
Merge git://git./linux/kernel/git/wim/linux-2.6-watchdog

* git://git.kernel.org/pub/scm/linux/kernel/git/wim/linux-2.6-watchdog:
  [WATCHDOG] bfin: fix max timeout calculation

14 years ago[WATCHDOG] bfin: fix max timeout calculation
Mike Frysinger [Tue, 16 Feb 2010 00:32:25 +0000 (19:32 -0500)]
[WATCHDOG] bfin: fix max timeout calculation

Relying on overflow/wrap around isn't exact because if you wrap far
enough, you get back to "valid" values.

Reported-by: Thorsten Pohlmann <pohlmann@tetronik.com>
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
Signed-off-by: Wim Van Sebroeck <wim@iguana.be>
Cc: stable <stable@kernel.org>
14 years agomm: Make copy_from_user() in migrate.c statically predictable
H. Peter Anvin [Fri, 19 Feb 2010 00:13:40 +0000 (16:13 -0800)]
mm: Make copy_from_user() in migrate.c statically predictable

x86-32 has had a static test for copy_on_user() overflow for a while.
This test currently fails in mm/migrate.c resulting in an
allyesconfig/allmodconfig build failure on x86-32:

In function â€˜copy_from_user’,
    inlined from â€˜do_pages_stat’ at
    /home/hpa/kernel/git/mm/migrate.c:1012:
/home/hpa/kernel/git/arch/x86/include/asm/uaccess_32.h:212: error:
    call to â€˜copy_from_user_overflow’ declared

Make the logic more explicit and therefore easier for gcc to
understand.

v2: rewrite the loop entirely using a more normal structure for a
    chunked-data loop (Linus Torvalds)

Reported-by: Len Brown <lenb@kernel.org>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Reviewed-and-Tested-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Arjan van de Ven <arjan@linux.kernel.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Christoph Lameter <cl@linux-foundation.org>
Cc: Hugh Dickins <hugh.dickins@tiscali.co.uk>
Cc: Rik van Riel <riel@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
14 years agoMerge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs-2.6
Linus Torvalds [Sun, 21 Feb 2010 00:58:03 +0000 (16:58 -0800)]
Merge branch 'for-linus' of git://git./linux/kernel/git/viro/vfs-2.6

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs-2.6:
  CacheFiles: Fix a race in cachefiles_delete_object() vs rename
  vfs: don't call ima_file_check() unconditionally in nfsd_open()
  fs: inode - remove 8 bytes of padding on 64bits allowing 1 more objects/slab under slub
  Switch proc/self to nd_set_link()
  fix LOOKUP_FOLLOW on automount "symlinks"

14 years agoMerge branch 'v4l_for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mchehab...
Linus Torvalds [Sun, 21 Feb 2010 00:56:09 +0000 (16:56 -0800)]
Merge branch 'v4l_for_linus' of git://git./linux/kernel/git/mchehab/linux-2.6

* 'v4l_for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mchehab/linux-2.6:
  V4L/DVB: bttv: Move I2C IR initialization
  V4L/DVB: Video : pwc : Fix regression in pwc_set_shutter_speed caused by bad  constant => sizeof conversion.
  soc-camera: mt9t112: modify exiting conditions from standby mode
  V4L/DVB: cxusb: Select all required frontend and tuner modules
  V4L/DVB: dvb: l64781.ko broken with gcc 4.5

14 years agoMerge branch 'omap-fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Sun, 21 Feb 2010 00:55:41 +0000 (16:55 -0800)]
Merge branch 'omap-fixes-for-linus' of git://git./linux/kernel/git/tmlind/linux-omap-2.6

* 'omap-fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tmlind/linux-omap-2.6:
  omap: Remove DEBUG_FS dependency for mux name checking

14 years agoMerge master.kernel.org:/home/rmk/linux-2.6-arm
Linus Torvalds [Sun, 21 Feb 2010 00:55:05 +0000 (16:55 -0800)]
Merge master.kernel.org:/home/rmk/linux-2.6-arm

* master.kernel.org:/home/rmk/linux-2.6-arm:
  ARM: allow alignment fault mode to be configured at kernel boot
  ARM: Update mach-types
  ARM: 5951/1: ARM: fix documentation of the PrimeCell bus
  ARM: 5950/1: ARM: Fix build error for arm1026ej-s processor
  MAINTAINERS: fix my e-mail and status for Gemini and FA526
  Gemini: wrong registers used to set reg_level in gpio_set_irq_type()
  ARM: 5944/1: scsi: fix timer setup in fas216.c
  ARM: 5938/1: ARM: L2: export outer_cache_fns

14 years agoARM: allow alignment fault mode to be configured at kernel boot
Russell King [Sat, 20 Feb 2010 16:13:29 +0000 (16:13 +0000)]
ARM: allow alignment fault mode to be configured at kernel boot

Some glibc versions intentionally create lots of alignment faults in
their gconv code, which if not fixed up, results in segfaults during
boot.  This can prevent systems booting properly.

There is no clear hard-configurable default for this; the desired
default depends on the nature of the userspace which is going to be
booted.

So, provide a way for the alignment fault handler to be configured via
the kernel command line.

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
14 years agoCacheFiles: Fix a race in cachefiles_delete_object() vs rename
David Howells [Fri, 19 Feb 2010 18:14:21 +0000 (18:14 +0000)]
CacheFiles: Fix a race in cachefiles_delete_object() vs rename

cachefiles_delete_object() can race with rename.  It gets the parent directory
of the object it's asked to delete, then locks it - but rename may have changed
the object's parent between the get and the completion of the lock.

However, if such a circumstance is detected, we abandon our attempt to delete
the object - since it's no longer in the index key path, it won't be seen
again by lookups of that key.  The assumption is that cachefilesd may have
culled it by renaming it to the graveyard for later destruction.

Signed-off-by: David Howells <dhowells@redhat.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
14 years agoARM: Update mach-types
Russell King [Sat, 20 Feb 2010 14:16:16 +0000 (14:16 +0000)]
ARM: Update mach-types

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
14 years agoARM: 5951/1: ARM: fix documentation of the PrimeCell bus
Linus Walleij [Sat, 20 Feb 2010 08:41:30 +0000 (09:41 +0100)]
ARM: 5951/1: ARM: fix documentation of the PrimeCell bus

This fixes the filepath encoded in <linux/amba/bus.h> and adds
some documentation as to what this bus really means.

Signed-off-by: Linus Walleij <linus.walleij@stericsson.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
14 years agoARM: 5950/1: ARM: Fix build error for arm1026ej-s processor
Abdoulaye Walsimou Gaye [Fri, 19 Feb 2010 11:47:14 +0000 (12:47 +0100)]
ARM: 5950/1: ARM: Fix build error for arm1026ej-s processor

This patch fix the below build error for arm1026ej-s processor (IntegratorCP/arm1026ej-s board).
  CC      init/main.o
In file included from include/linux/highmem.h:8,
                 from include/linux/pagemap.h:10,
                 from include/linux/mempolicy.h:62,
                 from init/main.c:52:
arch/arm/include/asm/cacheflush.h:134:2: error: #error Unknown cache maintainence model
make[1]: *** [init/main.o] Erreur 1
make: *** [init] Erreur 2

Signed-off-by: Abdoulaye Walsimou Gaye <walsimou@walsimou.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
14 years agoInput: polldev can cause crash in case when polling disabled
Samu Onkalo [Sat, 20 Feb 2010 07:17:58 +0000 (23:17 -0800)]
Input: polldev can cause crash in case when polling disabled

When polled input device is opened and closed and there are no other
users of polled device, the workqueue is created and destroyed in
every open / close operation. It is probable that at some point
dynamic allocation of internal parts of the workqueue cause changes to the
workqueue.

When a work is queued to the workqueue the work struct contains pointers
to the workqueue data. If the workqueue has been changed and the work
has never been queued to the new workqueue, work-struct contains pointers
to the non-existing workqueue. This will cause crash at the work
cancellation during device close since cancellation of a work assumes
that the workqueue exists.

To prevent that, work struct is cleaned up at device close. This keeps
work struct clean for the next use.

Signed-off-by: Samu Onkalo <samu.p.onkalo@nokia.com>
Signed-off-by: Dmitry Torokhov <dtor@mail.ru>
14 years agovfs: don't call ima_file_check() unconditionally in nfsd_open()
Chuck Ebbert [Mon, 15 Feb 2010 23:07:39 +0000 (18:07 -0500)]
vfs: don't call ima_file_check() unconditionally in nfsd_open()

commit 1e41568d7378d1ba8c64ba137b9ddd00b59f893a ("Take ima_path_check()
in nfsd past dentry_open() in nfsd_open()") moved this code back to its
original location but missed the "else".

Signed-off-by: Chuck Ebbert <cebbert@redhat.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
14 years agoMerge git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi-rc-fixes-2.6
Linus Torvalds [Sat, 20 Feb 2010 03:35:20 +0000 (19:35 -0800)]
Merge git://git./linux/kernel/git/jejb/scsi-rc-fixes-2.6

* git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi-rc-fixes-2.6:
  [SCSI] megaraid_sas: fix for 32bit apps
  [SCSI] fcoe: Only rmmod fcoe.ko if there are no active connections
  [SCSI] libfcoe: Send port LKA every FIP_VN_KA_PERIOD secs.
  [SCSI] libfc: Don't assume response request present.
  [SCSI] libfc: Fix e_d_tov ns -> ms scaling factor in PLOGI response.
  [SCSI] libfc: call ddp setup for only FCP reads to avoid accessing junk fsp pointer
  [SCSI] iscsi_tcp regression: remove bogus warn on in write path

14 years agoMerge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-2.6
Linus Torvalds [Sat, 20 Feb 2010 03:34:56 +0000 (19:34 -0800)]
Merge git://git./linux/kernel/git/davem/net-2.6

* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-2.6:
  sfc: SFE4002/SFN4112F: Widen temperature and voltage tolerances
  sfc: Fix sign of efx_mcdi_poll_reboot() error in efx_mcdi_poll()
  net-sysfs: Use rtnl_trylock in wireless sysfs methods.
  net: Fix sysctl restarts...

14 years agoMerge git://git.kernel.org/pub/scm/linux/kernel/git/davem/sparc-2.6
Linus Torvalds [Sat, 20 Feb 2010 03:34:40 +0000 (19:34 -0800)]
Merge git://git./linux/kernel/git/davem/sparc-2.6

* git://git.kernel.org/pub/scm/linux/kernel/git/davem/sparc-2.6:
  sparc64: Fix sun4u execute bit check in TSB I-TLB load.
  sparc: Fix incorrect comparison in of_bus_ambapp_match()
  sparc64: Sync of_create_pci_dev() with drivers/pci/probe.c changes.
  sparc64: Tighten checks in kstack_valid().

14 years agoMerge branch 'drm-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/airlied...
Linus Torvalds [Sat, 20 Feb 2010 03:33:51 +0000 (19:33 -0800)]
Merge branch 'drm-linus' of git://git./linux/kernel/git/airlied/drm-2.6

* 'drm-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/airlied/drm-2.6:
  drm/radeon: bump the UMS driver version number to indicate rv740 fix
  drm/radeon/kms: free fence IB if it wasn't emited at IB free time
  drm/ttm: fix caching problem on non-PAT systems.
  drm/radeon/rv740: fix backend setup
  drm/radeon/kms: fix shared ddc detection
  drm/radeon/kms/rs600: add connector quirk
  vgaarb: fix "target=default" passing

14 years agosfc: SFE4002/SFN4112F: Widen temperature and voltage tolerances
Ben Hutchings [Fri, 19 Feb 2010 13:34:03 +0000 (13:34 +0000)]
sfc: SFE4002/SFN4112F: Widen temperature and voltage tolerances

The temperature and voltage limits currently set on these boards are
too conservative and will cause the driver to stop the net device
erroneously in some systems.

Based on a review of the chip datasheets and advice from the designer
of these boards:

- Raise the maximum board temperatures to the specified maximum ambient
  temperatures for their PHYs plus the expected temperature bias of the
  board
- Raise the maximum controller temperature to 90 degrees
- Lower the minimum temperatures to 0 degrees
- Widen the voltage tolerances to at least +/- 10%

Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agosfc: Fix sign of efx_mcdi_poll_reboot() error in efx_mcdi_poll()
Ben Hutchings [Fri, 19 Feb 2010 13:29:27 +0000 (13:29 +0000)]
sfc: Fix sign of efx_mcdi_poll_reboot() error in efx_mcdi_poll()

efx_mcdi_poll() uses positive error numbers, matching the MCDI
protocol.  It must negate the result of efx_mcdi_poll_reboot() which
returns the usual negative error numbers.

Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agonet-sysfs: Use rtnl_trylock in wireless sysfs methods.
Eric W. Biederman [Fri, 19 Feb 2010 13:23:47 +0000 (13:23 +0000)]
net-sysfs: Use rtnl_trylock in wireless sysfs methods.

The wireless sysfs methods like the rest of the networking sysfs
methods are removed with the rtnl_lock held and block until
the existing methods stop executing.  So use rtnl_trylock
and restart_syscall so that the code continues to work.

Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agonet: Fix sysctl restarts...
Eric W. Biederman [Fri, 19 Feb 2010 13:22:59 +0000 (13:22 +0000)]
net: Fix sysctl restarts...

Yuck.  It turns out that when we restart sysctls we were restarting
with the values already changed.  Which unfortunately meant that
the second time through we thought there was no change and skipped
all kinds of work, despite the fact that there was indeed a change.

I have fixed this the simplest way possible by restoring the changed
values when we restart the sysctl write.

One of my coworkers spotted this bug when after disabling forwarding
on an interface pings were still forwarded.

Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agosparc64: Fix sun4u execute bit check in TSB I-TLB load.
David S. Miller [Fri, 19 Feb 2010 23:19:52 +0000 (15:19 -0800)]
sparc64: Fix sun4u execute bit check in TSB I-TLB load.

Thanks to testcase and report from Brad Spengler:

--------------------
#include <stdio.h>

typedef int (* _wee)(void);

int main(void)
{
        char buf[8] = { '\x81', '\xc7', '\xe0', '\x08', '\x81', '\xe8',
                        '\x00', '\x00' };
        _wee wee;
        printf("%p\n", &buf);
        wee = (_wee)&buf;
        wee();

        return 0;
}
--------------------

TSB I-tlb load code tries to use andcc to check the _PAGE_EXEC_4U bit,
but that's bit 12 so it gets sign extended all the way up to bit 63
and the test nearly always passes as a result.

Use sethi to fix the bug.

Signed-off-by: David S. Miller <davem@davemloft.net>
14 years agodrm/radeon: bump the UMS driver version number to indicate rv740 fix
Dave Airlie [Fri, 19 Feb 2010 23:17:18 +0000 (09:17 +1000)]
drm/radeon: bump the UMS driver version number to indicate rv740 fix

This lets UMS userspace know the rv740 fix is in. For KMS we can
consider the kernel release to be the v2.0.0 release so we don't need the
bump there.

Signed-off-by: Dave Airlie <airlied@redhat.com>
14 years agodrm/radeon/kms: free fence IB if it wasn't emited at IB free time
Jerome Glisse [Thu, 18 Feb 2010 13:13:29 +0000 (13:13 +0000)]
drm/radeon/kms: free fence IB if it wasn't emited at IB free time

If at IB free time fence wasn't emited that means the IB wasn't
scheduled because an error occured somewhere, thus we can free
then fence and mark the IB as free.

Signed-off-by: Jerome Glisse <jglisse@redhat.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
14 years agodrm/ttm: fix caching problem on non-PAT systems.
Francisco Jerez [Fri, 19 Feb 2010 21:30:15 +0000 (07:30 +1000)]
drm/ttm: fix caching problem on non-PAT systems.

http://bugzilla.kernel.org/show_bug.cgi?id=15328

This fixes a serious regression on AGP/non-PAT systems, where
pages were ending up in the wrong state and slowing down the
whole system.

[airlied: taken this from the bug as the other option is to revert
the change which caused it].

Tested-by: John W. Linville (in bug).
Signed-off-by: Dave Airlie <airlied@redhat.com>
14 years agodrm/radeon/rv740: fix backend setup
Alex Deucher [Fri, 19 Feb 2010 21:07:02 +0000 (16:07 -0500)]
drm/radeon/rv740: fix backend setup

This patch fixes occlusion queries and rendering errors
on rv740 boards. Hardcoding the backend map is not an optimal
solution, but a better fix is being worked on.

Signed-off-by: Alex Deucher <alexdeucher@gmail.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
14 years agodrm/radeon/kms: fix shared ddc detection
Alex Deucher [Fri, 19 Feb 2010 07:13:56 +0000 (02:13 -0500)]
drm/radeon/kms: fix shared ddc detection

Just compare the i2c id since the i2c structs
may be slighly different.

Fixes fdo bug 26616.

Signed-off-by: Alex Deucher <alexdeucher@gmail.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
14 years agodrm/radeon/kms/rs600: add connector quirk
Alex Deucher [Thu, 18 Feb 2010 19:14:58 +0000 (14:14 -0500)]
drm/radeon/kms/rs600: add connector quirk

rs600 board lists DVI port as HDMI.

Fixes fdo bug 26605

Signed-off-by: Alex Deucher <alexdeucher@gmail.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
14 years agovgaarb: fix "target=default" passing
Kyle McMartin [Tue, 16 Feb 2010 21:18:37 +0000 (16:18 -0500)]
vgaarb: fix "target=default" passing

Commit 77c1ff3982c6b36961725dd19e872a1c07df7f3b fixed the userspace
pointer dereference, but introduced another bug pointed out by Eugene Teo
in RH bug #564264. Instead of comparing the point we were at in the string,
we instead compared the beginning of the string to "default".

Signed-off-by: Kyle McMartin <kyle@redhat.com>
Reported-by: Eugene Teo <eteo@redhat.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
14 years ago[SCSI] megaraid_sas: fix for 32bit apps
Tomas Henzl [Thu, 11 Feb 2010 17:01:50 +0000 (18:01 +0100)]
[SCSI] megaraid_sas: fix for 32bit apps

It looks like this patch -

commit 7b2519afa1abd1b9f63aa1e90879307842422dae
Author: Yang, Bo <Bo.Yang@lsi.com>
Date:   Tue Oct 6 14:52:20 2009 -0600

    [SCSI] megaraid_sas: fix 64 bit sense pointer truncation

has caused a problem for 32bit programs with 64bit os -

http://bugzilla.kernel.org/show_bug.cgi?id=15001

fix by converting the user space 32bit pointer to a 64 bit one when
needed.

[jejb: fix up some 64 bit warnings]
Signed-off-by: Tomas Henzl <thenzl@redhat.com>
Cc: Bo Yang <Bo.Yang@lsi.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
14 years agohw-breakpoint: Keep track of dr7 local enable bits
Frederic Weisbecker [Thu, 18 Feb 2010 17:24:18 +0000 (18:24 +0100)]
hw-breakpoint: Keep track of dr7 local enable bits

When the user enables breakpoints through dr7, he can choose
between "local" or "global" enable bits but given how linux is
implemented, both have the same effect.

That said we don't keep track how the user enabled the breakpoints
so when the user requests the dr7 value, we only translate the
"enabled" status using the global enabled bits. It means that if
the user enabled a breakpoint using the local enabled bit, reading
back dr7 will set the global bit and clear the local one.

Apps like Wine expect a full dr7 POKEUSER/PEEKUSER match for emulated
softwares that implement old reverse engineering protection schemes.

We fix that by keeping track of the whole dr7 value given by the user
in the thread structure to drop this bug. We'll think about
something more proper later.

This fixes a 2.6.32 - 2.6.33-x ptrace regression.

Reported-and-tested-by: Michael Stefaniuc <mstefani@redhat.com>
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
Acked-by: K.Prasad <prasad@linux.vnet.ibm.com>
Cc: Alan Stern <stern@rowland.harvard.edu>
Cc: Maneesh Soni <maneesh@linux.vnet.ibm.com>
Cc: Alexandre Julliard <julliard@winehq.org>
Cc: Rafael J. Wysocki <rjw@sisk.pl>
Cc: Maciej Rutecki <maciej.rutecki@gmail.com>
14 years agohw-breakpoints: Accept breakpoints on NULL address
Frederic Weisbecker [Thu, 18 Feb 2010 15:00:59 +0000 (16:00 +0100)]
hw-breakpoints: Accept breakpoints on NULL address

Before we had a generic breakpoint API, ptrace was accepting
breakpoints on NULL address in x86. The new API refuse them,
without given strong reasons. We need to follow the previous
behaviour as some userspace apps like Wine need such NULL
breakpoints to ensure old emulated software protections
are still working.

This fixes a 2.6.32 - 2.6.33-x ptrace regression.

Reported-and-tested-by: Michael Stefaniuc <mstefani@redhat.com>
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
Acked-by: K.Prasad <prasad@linux.vnet.ibm.com>
Acked-by: Roland McGrath <roland@redhat.com>
Cc: Alan Stern <stern@rowland.harvard.edu>
Cc: Maneesh Soni <maneesh@linux.vnet.ibm.com>
Cc: Alexandre Julliard <julliard@winehq.org>
Cc: Rafael J. Wysocki <rjw@sisk.pl>
Cc: Maciej Rutecki <maciej.rutecki@gmail.com>
14 years agofs: inode - remove 8 bytes of padding on 64bits allowing 1 more objects/slab under...
Richard Kennedy [Mon, 15 Feb 2010 11:16:11 +0000 (11:16 +0000)]
fs: inode - remove 8 bytes of padding on 64bits allowing 1 more objects/slab under slub

This removes 8 bytes of padding from struct inode on 64bit builds, and
so allows 1 more object/slab in the inode_cache when using slub.

Signed-off-by: Richard Kennedy <richard@rsk.demon.co.uk>
----
patch against 2.6.33-rc8
compiled & tested on x86_64 AMDX2

I've been running this patch for over a week with no obvious problems
regards
Richard
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
14 years agoSwitch proc/self to nd_set_link()
Al Viro [Thu, 14 Jan 2010 06:03:28 +0000 (01:03 -0500)]
Switch proc/self to nd_set_link()

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
14 years agofix LOOKUP_FOLLOW on automount "symlinks"
Al Viro [Tue, 16 Feb 2010 18:09:36 +0000 (18:09 +0000)]
fix LOOKUP_FOLLOW on automount "symlinks"

Make sure that automount "symlinks" are followed regardless of LOOKUP_FOLLOW;
it should have no effect on them.

Cc: stable@kernel.org
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
14 years agoV4L/DVB: bttv: Move I2C IR initialization
Jean Delvare [Fri, 19 Feb 2010 03:18:41 +0000 (00:18 -0300)]
V4L/DVB: bttv: Move I2C IR initialization

Move I2C IR initialization from just after I2C bus setup to right
before non-I2C IR initialization. This avoids the case where an I2C IR
device is blocking audio support (at least the PV951 suffers from
this). It is also more logical to group IR support together,
regardless of the connectivity.

This fixes bug #15184:
http://bugzilla.kernel.org/show_bug.cgi?id=15184

Signed-off-by: Jean Delvare <khali@linux-fr.org>
CC: stable@kernel.org
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
14 years agoV4L/DVB: Video : pwc : Fix regression in pwc_set_shutter_speed caused by bad constan...
Martin Fuzzey [Thu, 11 Feb 2010 13:50:31 +0000 (10:50 -0300)]
V4L/DVB: Video : pwc : Fix regression in pwc_set_shutter_speed caused by bad  constant => sizeof conversion.

Regression was caused by my commit 6b35ca0d3d586b8ecb8396821af21186e20afaf0
which determined message size using sizeof rather than hardcoded constants.

Unfortunately pwc_set_shutter_speed reuses a 2 byte buffer for a one byte
message too so the sizeof was bogus in this case.

All other uses of sizeof checked and are ok.

Acked-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Cc: stable@kernel.org
Signed-off-by: Martin Fuzzey <mfuzzey@gmail.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
14 years agosoc-camera: mt9t112: modify exiting conditions from standby mode
Kuninori Morimoto [Tue, 2 Feb 2010 04:17:54 +0000 (13:17 +0900)]
soc-camera: mt9t112: modify exiting conditions from standby mode

This polling is needed if camera is in standby mode, but current exiting
condition is inverted.

Signed-off-by: Kuninori Morimoto <morimoto.kuninori@renesas.com>
Signed-off-by: Guennadi Liakhovetski <g.liakhovetski@gmx.de>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
14 years agoV4L/DVB: cxusb: Select all required frontend and tuner modules
Ben Hutchings [Thu, 11 Feb 2010 02:57:17 +0000 (23:57 -0300)]
V4L/DVB: cxusb: Select all required frontend and tuner modules

cxusb uses the atbm8830 and lgs8gxx (not lgs8gl5) frontends and the
max2165 tuner, so it needs to select them.

Signed-off-by: Ben Hutchings <ben@decadent.org.uk>
Cc: stable@kernel.org
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
14 years agoV4L/DVB: dvb: l64781.ko broken with gcc 4.5
Richard Guenther [Tue, 9 Feb 2010 23:16:03 +0000 (20:16 -0300)]
V4L/DVB: dvb: l64781.ko broken with gcc 4.5

I'm trying to fix it on the GCC side (PR43007), but the module is
quite stupid in using ULL constants to operate on u32 values:

static int apply_frontend_param (struct dvb_frontend* fe, struct
dvb_frontend_parameters *param)
{
...
 static const u32 ppm = 8000;
 u32 spi_bias;
...

 spi_bias *= 1000ULL;
 spi_bias /= 1000ULL + ppm/1000;

which causes current GCC 4.5 to emit calls to __udivdi3 for i?86 again.

This patch fixes this issue.

Signed-off-by: Richard Guenther <rguenther@suse.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
CC: stable@kernel.org
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>