UBUNTU: debian -- pass the correct flavour name when checking configs

Signed-off-by: Andy Whitcroft <apw@canonical.com>

UBUNTU: [Config] enforcer -- ensure we never enable CONFIG_IMA

Signed-off-by: Andy Whitcroft <apw@canonical.com>

UBUNTU: [Config] Disable intel_idle for -virtual kernels

BugLink: http://bugs.launchpad.net/bugs/651370

EC2 Xen does not properly handle the mwait instruction used by intel_idle
resulting in an invalid opcode when an instance is launched on newer
intel hardware.

Signed-off-by: John Johansen <john.johansen@canonical.com>
Signed-off-by: Andy Whitcroft <apw@canonical.com>

UBUNTU: SAUCE: nx-emu: use upstream ASLR when possible

Only perform ASCII-armor ASLR when nx-emulation is happening.

Signed-off-by: Kees Cook <kees.cook@canonical.com>
Signed-off-by: Andy Whitcroft <apw@canonical.com>

exec-randomization: brk away from exec rand area

This is a fix for the NX emulation patch to force the brk area well
outside of the exec randomization area to avoid future allocation or brk
growth collisions. Normally this isn't a problem, except when the text
region has been loaded from a PIE binary and the CS limit can't be put
just above bss.

A test-case that will show failures without this patch can be found here:
http://bazaar.launchpad.net/~ubuntu-bugcontrol/qa-regression-testing/master/annotate/head%3A/scripts/kernel-aslr-collisions/explode-brk.c

Signed-off-by: Kees Cook <kees.cook@canonical.com>
Signed-off-by: Andy Whitcroft <apw@canonical.com>

mmap randomization for executable mappings on 32-bit

This code is originally from Ingo, with some later rebasing and
fixes to respect all the randomization-disabling knobs.

It provides a new address randomization algorithm that's preferable
for executable mappings in 32-bit processes.

Signed-off-by: Roland McGrath <roland@redhat.com>
Signed-off-by: Kees Cook <kees.cook@canonical.com>
Signed-off-by: Andy Whitcroft <apw@canonical.com>

nx-emu: standardize boottime message prefix

The state of NX in the booted kernel is easier to find in dmesg if
all the possible modes start with the same bit of text. There's no
good reason to have the NX-emulation message have a different format.

Signed-off-by: Kees Cook <kees.cook@canonical.com>
Signed-off-by: Andy Whitcroft <apw@canonical.com>

nx-emu: drop exec-shield sysctl, merge with disable_nx

The "exec-shield" sysctl doesn't really make much sense any more, and
the functionality is better merged with the existing "disable_nx"
kernel boot option.

Signed-off-by: Kees Cook <kees.cook@canonical.com>
Signed-off-by: Andy Whitcroft <apw@canonical.com>

i386: NX emulation

This is old code with some cruft, all originally by Ingo with much
later rebasing by Fedora folks and at least one arcane fix by Roland
a few years ago.

Signed-off-by: Roland McGrath <roland@redhat.com>
Signed-off-by: Kees Cook <kees.cook@canonical.com>
Signed-off-by: Andy Whitcroft <apw@canonical.com>

UBUNTU: Bump ABI

Ignore: yes
Signed-off-by: Andy Whitcroft <apw@canonical.com>

UBUNTU: Start new release

Ignore: yes
Signed-off-by: Andy Whitcroft <apw@canonical.com>

UBUNTU: [Config] Use correct be2iscsi module name in d-i/modules/scsi-modules

BugLink: http://bugs.launchpad.net/bugs/628776

Signed-off-by: Tim Gardner <tim.gardner@canonical.com>

UBUNTU: Ubuntu-2.6.37-2.10

Signed-off-by: Andy Whitcroft <apw@canonical.com>

UBUNTU: [Config] CONFIG_GPIO_VX855=n for armel FTBS

Signed-off-by: Tim Gardner <tim.gardner@canonical.com>
Signed-off-by: Andy Whitcroft <apw@canonical.com>

UBUNTU: Start new release

Ignore: yes
Signed-off-by: Andy Whitcroft <apw@canonical.com>

UBUNTU: [Config] CONFIG_GPIO_PCH=n for armel FTBS

Signed-off-by: Tim Gardner <tim.gardner@canonical.com>
Signed-off-by: Andy Whitcroft <apw@canonical.com>

UBUNTU: Ubuntu-2.6.37-2.9

Signed-off-by: Andy Whitcroft <apw@canonical.com>

UBUNTU: SAUCE: fix documentation strings for struct input_keymap_entry

Signed-off-by: Andy Whitcroft <apw@canonical.com>

UBUNTU: bump master version number to match contained kernel

Signed-off-by: Andy Whitcroft <apw@canonical.com>

UBUNTU: Start new release

Ignore: yes
Signed-off-by: Andy Whitcroft <apw@canonical.com>

UBUNTU: [Config] move powerpc over from ports to distro

Signed-off-by: Andy Whitcroft <apw@canonical.com>

UBUNTU: config -- fix genportsconfig

Signed-off-by: Andy Whitcroft <apw@canonical.com>

UBUNTU: Ubuntu-2.6.36-2.8

Signed-off-by: Tim Gardner <tim.gardner@canonical.com>

UBUNTU: SAUCE: AppArmor: Fix unpack of network tables.

The unpacking of network rules, unpacks 1 more rule than it should.  It
should drop all rules with network types AF_MAX or greater.

Fix suggested by Tetsuo Handa in
https://lists.ubuntu.com/archives/kernel-team/2010-November/013327.html

Reported-by: Tetsuo Handa <from-ubuntu@I-love.SAKURA.ne.jp>
Signed-off-by: John Johansen <john.johansen@canonical.com>
Signed-off-by: Tim Gardner <tim.gardner@canonical.com>

UBUNTU: [Config] CONFIG_DRM_NOUVEAU=n for armel FTBS

Signed-off-by: Tim Gardner <tim.gardner@canonical.com>

UBUNTU: [Config] CONFIG_LINE6_USB=n for armel FTBS

Signed-off-by: Tim Gardner <tim.gardner@canonical.com>

UBUNTU: [Config] CONFIG_SENSORS_AK8975=n for armel FTBS

Signed-off-by: Tim Gardner <tim.gardner@canonical.com>

UBUNTU: [Config] CONFIG_I2C_I801=n for armel FTBS

Signed-off-by: Tim Gardner <tim.gardner@canonical.com>

UBUNTU: [Config] CONFIG_GPIO_VX855=n for armel FTBS

Signed-off-by: Tim Gardner <tim.gardner@canonical.com>

UBUNTU: [Config] CONFIG_GPIO_PCH=n for armel FTBS

Signed-off-by: Tim Gardner <tim.gardner@canonical.com>

UBUNTU: [Config] CONFIG_SECURITY_APPARMOR_COMPAT_24=y

Signed-off-by: Andy Whitcroft <apw@canonical.com>
Signed-off-by: Tim Gardner <tim.gardner@canonical.com>

UBUNTU: updateportsconfigs following rebase to v2.6.37-rc1

Signed-off-by: Andy Whitcroft <apw@canonical.com>
Signed-off-by: Tim Gardner <tim.gardner@canonical.com>

UBUNTU: [Config] CONFIG_SECURITY_APPARMOR_COMPAT_24=y

Signed-off-by: Tim Gardner <tim.gardner@canonical.com>

AppArmor: compatibility patch for v5 interface

Signed-off-by: John Johansen <john.johansen@canonical.com>
Signed-off-by: Tim Gardner <tim.gardner@canonical.com>

AppArmor: compatibility patch for v5 network controll

Add compatibility for v5 network rules.

Signed-off-by: John Johansen <john.johansen@canonical.com>
Signed-off-by: Tim Gardner <tim.gardner@canonical.com>

UBUNTU: [Config] Clean intermediate directories to make builds repeatable

Signed-off-by: Tim Gardner <tim.gardner@canonical.com>

UBUNTU: [Config] Updated to new firewire module names.

Signed-off-by: Tim Gardner <tim.gardner@canonical.com>

UBUNTU: [Config] Drop bnx2 firmware from udeb until the upstream Makefile issue is resolved

Signed-off-by: Tim Gardner <tim.gardner@canonical.com>

UBUNTU: udeb: updated bnx2 firmware file names

Signed-off-by: Tim Gardner <tim.gardner@canonical.com>

UBUNTU: [Config]: CONFIG_SOLO6X10=n for FTBS

Signed-off-by: Tim Gardner <tim.gardner@canonical.com>

UBUNTU: [Config]: CONFIG_ATH6K_LEGACY=n for FTBS

Signed-off-by: Tim Gardner <tim.gardner@canonical.com>

UBUNTU: [Config]: TWL4030_CORE=n

Signed-off-by: Tim Gardner <tim.gardner@canonical.com>

UBUNTU: [Config]: fix changed CONFIG_SYSFS_DEPRECATED_V2 enforcement rules

Signed-off-by: Tim Gardner <tim.gardner@canonical.com>

UBUNTU: [Config]: updateconfigs after rebase to 2.6.37-rc1

Signed-off-by: Tim Gardner <tim.gardner@canonical.com>

UBUNTU: [Config] Note dropped SAUCE patches from 2.6.36 to 2.6.37

Signed-off-by: Tim Gardner <tim.gardner@canonical.com>

UBUNTU: Start new release

Ignore: yes
Signed-off-by: Tim Gardner <tim.gardner@canonical.com>

UBUNTU: Ubuntu-2.6.36-1.7

Signed-off-by: Andy Whitcroft <apw@canonical.com>

UBUNTU: [Config] update configs following rebase to v2.6.36 final

Signed-off-by: Andy Whitcroft <apw@canonical.com>

UBUNTU: rebased to v2.6.36 final

Signed-off-by: Andy Whitcroft <apw@canonical.com>

UBUNTU: Start new release

Ignore: yes
Signed-off-by: Andy Whitcroft <apw@canonical.com>

UBUNTU: Ubuntu-2.6.36-1.6

Signed-off-by: Andy Whitcroft <apw@canonical.com>

UBUNTU: Start new release

Ignore: yes
Signed-off-by: Andy Whitcroft <apw@canonical.com>

UBUNTU: Ubuntu-2.6.36-1.5

Signed-off-by: Andy Whitcroft <apw@canonical.com>

UBUNTU: config -- add new config mode 'dumpconfigs'

Add a new configuration mode 'dumpconfigs' which dumps out all of the
configuration questions and their current values.  This is consumed by
external scripting to acertain the current options, their types,
and default and current values.

Signed-off-by: Andy Whitcroft <apw@canonical.com>

UBUNTU: config -- simplify the kernelconfig interface

In preparation for adding another kernelconfig interface, simply
the current interface such that the real command is passed through.
This reduces the number and complexity of the make rules while maintaining
a single mapping in kernelconfig.

Signed-off-by: Andy Whitcroft <apw@canonical.com>

UBUNTU: Bump ABI

Ignore: yes
Signed-off-by: Andy Whitcroft <apw@canonical.com>

UBUNTU: updateportsconfigs following rebase to v2.6.36-rc8

Signed-off-by: Andy Whitcroft <apw@canonical.com>

UBUNTU: updateconfigs following rebase to v2.6.36-rc8

Signed-off-by: Andy Whitcroft <apw@canonical.com>

UBUNTU: rebase to v2.6.36-rc8

Signed-off-by: Andy Whitcroft <apw@canonical.com>

UBUNTU: Start new release

Ignore: yes
Signed-off-by: Andy Whitcroft <apw@canonical.com>

UBUNTU: Simplify the use of CROSS_COMPILER

Signed-off-by: Tim Gardner <tim.gardner@canonical.com>

UBUNTU: Ubuntu-2.6.36-0.4

Signed-off-by: Andy Whitcroft <apw@canonical.com>

UBUNTU: Start new release

Ignore: yes
Signed-off-by: Andy Whitcroft <apw@canonical.com>

UBUNTU: Ubuntu-2.6.36-0.3

Signed-off-by: Andy Whitcroft <apw@canonical.com>

UBUNTU: [Config] disable CONFIG_SCSI_QLA_ISCSI to fix FTBS on powerpc

Signed-off-by: Andy Whitcroft <apw@canonical.com>

UBUNTU: Start new release

Ignore: yes
Signed-off-by: Andy Whitcroft <apw@canonical.com>

UBUNTU: Ubuntu-2.6.36-0.2

Signed-off-by: Andy Whitcroft <apw@canonical.com>

UBUNTU: Start new release

Ignore: yes
Signed-off-by: Andy Whitcroft <apw@canonical.com>

UBUNTU: [Config] updateportsconfigs following rebase to 2.6.36-rc7

Signed-off-by: Andy Whitcroft <apw@canonical.com>

UBUNTU: Ubuntu-2.6.36-0.1

Signed-off-by: Andy Whitcroft <apw@canonical.com>

UBUNTU: [Config] armel -- drop omap flavour

We are expecting to pick up the main arm flavours externally so drop
all but versatile from the main kernel for the moment.

Signed-off-by: Andy Whitcroft <apw@canonical.com>

UBUNTU: [Config] enforcer -- ensure CONFIG_INIT_PASS_ALL_PARAMS is y

Signed-off-by: Andy Whitcroft <apw@canonical.com>

UBUNTU: [Config]: updateconfigs after adding brcm80211

Signed-off-by: Tim Gardner <tim.gardner@canonical.com>

UBUNTU: [Upstream] drivers/serial/mfd.c: Fix ARM compile error

drivers/serial/mfd.c: In function 'port_show_regs':
drivers/serial/mfd.c:144: error: implicit declaration of function 'kzalloc'

Signed-off-by: Tim Gardner <tim.gardner@canonical.com>

UBUNTU: [Config] CONFIG_SCSI_QLA_ISCSI=n for armel

It seems that readq/witeq were not defined for the ARM arch.

Signed-off-by: Tim Gardner <tim.gardner@canonical.com>

UBUNTU: [Config] Add support for cross compiling armel

The best way to cross compile is by using dpkg-buildpackage, e.g.,

dpkg-buildpackage -B -aarmel

You'll likely have problems cross compiling the tools unless you've
built and installed some prerequisites, but thats your problem. Work
around the issue by deleting the tools directorty or specify do_tools=false.

Signed-off-by: Tim Gardner <tim.gardner@canonical.com>

UBUNTU: reduce disk usage during buildd builds

BugLink: http://bugs.launchpad.net/bugs/645653

For each flavour build we keep all of intermediate trees intact.
This allows simple incremental updates in a development environment but
leads to very large disk footprints on the build.  For Maverick with three
flavours we are seeing 19GB for i386 and 25GB for amd64.  Anything over
15GB is likely to run the buildds out of disk and fail the build.

When building on a buildd (in full_build=true mode) clean out
the main build tree and the package tree for the debugging deb.
This reduces consumption to more like 10GB maximum regardless of
the flavour count.

Signed-off-by: Andy Whitcroft <apw@canonical.com>

UBUNTU: [Config] Disable aufs, dmraid-4.5, ndis-wrapper

Signed-off-by: Tim Gardner <tim.gardner@canonical.com>

UBUNTU: more dropped patches

Signed-off-by: Tim Gardner <rtg@tyler.taipei>

UBUNTU: Added dropped patch list

Signed-off-by: Tim Gardner <tim.gardner@canonical.com>

UBUNTU: SAUCE: Nouveau: Disable acceleration on MacBook Pros

BugLink: http://bugs.launchpad.net/bugs/546393

Upstream obviously wants to fix acceleration rather than disable it.
For Lucid however we need these cards to boot more than we need them to
be fast.

With acceleration enabled these cards hang during boot.

Signed-off-by: Christopher James Halse Rogers <raof@ubuntu.com>
Acked-by: Andy Whitcroft <apw@canonical.com>
Signed-off-by: Andy Whitcroft <apw@canonical.com>
Signed-off-by: Leann Ogasawara <leann.ogasawara@canonical.com>

UBUNTU: SAUCE: Nouveau: Add quirk framework to disable acceleration

Add support for Nouveau acceleration quirks.

BugLink: http://bugs.launchpad.net/bugs/546393
BugLink: http://bugs.launchpad.net/bugs/544088

Signed-off-by: Christopher James Halse Rogers <raof@ubuntu.com>
Acked-by: Andy Whitcroft <apw@canonical.com>
Signed-off-by: Andy Whitcroft <apw@canonical.com>
Signed-off-by: Leann Ogasawara <leann.ogasawara@canonical.com>

UBUNTU: SAUCE: Improve Amazon EBS performance for EC2

OriginalAuthor: Amazona from Ben Howard <behoward@amazon.com>
BugLink: http://bugs.launchpad.net/bugs/634316

The pv-ops kernel suffers from poor performance when using Amazon's
Elastic block storage (EBS).  This patch from Amazon improves pv-ops
kernel performance, and has not exhibited any regressions.

Signed-off-by: John Johansen <john.johansen@canonical.com>
Signed-off-by: Leann Ogasawara <leann.ogasawara@canonical.com>

UBUNTU: SAUCE: hid: ntrig: zero-initialize ntrig struct

Setting zeros using kzalloc instead.

Has been tested succesfully on Dell Studio 17, Dell XT2, HP TX2 and
Lenovo T410s. Not likely to go upstream in its present form. However,
it does help constitue a tuning-free driver which works well together
with Unity, and as such are vital to the MT push.

Signed-off-by: Henrik Rydberg <rydberg@euromail.se>
Acked-by: Chase Douglas <chase.douglas@canonical.com>
Signed-off-by: Leann Ogasawara <leann.ogasawara@canonical.com>

UBUNTU: SAUCE: hid: ntrig: Correct logic for quirks

The driver data is a bitmask, check for correct bits.

Has been tested succesfully on Dell Studio 17, Dell XT2, HP TX2 and
Lenovo T410s. Not likely to go upstream in its present form. However,
it does help constitue a tuning-free driver which works well together
with Unity, and as such are vital to the MT push.

Signed-off-by: Henrik Rydberg <rydberg@euromail.se>
Acked-by: Chase Douglas <chase.douglas@canonical.com>
Signed-off-by: Leann Ogasawara <leann.ogasawara@canonical.com>

UBUNTU: SAUCE: hid: ntrig: Remove unused device ids

The current set of devices (Dell Studio 17, Dell XT2, HP TX2, Lenovo T810s)
all use the same device id (1b96:0001). However, the driver defines a
range of device ids. Remove these and reintroduce when needed, in
anticipation that new devices will not work with the current driver
anyways.

Has been tested succesfully on Dell Studio 17, Dell XT2, HP TX2 and
Lenovo T410s. Not likely to go upstream in its present form. However,
it does help constitue a tuning-free driver which works well together
with Unity, and as such are vital to the MT push.

Signed-off-by: Henrik Rydberg <rydberg@euromail.se>
Acked-by: Chase Douglas <chase.douglas@canonical.com>
Signed-off-by: Leann Ogasawara <leann.ogasawara@canonical.com>

UBUNTU: SAUCE: ndiswrapper: Initialize buffer index and check its value

BugLink: http://bugs.launchpad.net/bugs/613796

The following commit propagated a change to ndiswrapper which originated
from upstream to convert multicast list to list_head.

  commit e2fce7c6af6e1bcd2f577304a5200738d076d98f
  Author: Leann Ogasawara <leann.ogasawara@canonical.com>
  Date:   Thu Jun 3 13:34:15 2010 -0700

    UBUNTU: SAUCE: ndiswrapper: convert multicast list to list_head

In making the change we accidentally forgot to carry forward
initialization of a buffer index and properly check its value.  Add this
missing functionality which previously existed.

We should squash this patch with commit e2fce7c6a upon the next rebase.

Acked-by: Tim Gardner <tim.gardner@canonical.com>
Signed-off-by: Leann Ogasawara <leann.ogasawara@canonical.com>

UBUNTU: SAUCE: fix compilation warning when CONFIG_SECURITY is not set

Function prototype differs when CONFIG_SECURITY is not set leading to
a warning:

include/linux/security.h: In function ‘security_task_free’:
include/linux/security.h:2244: warning: no return statement in function returning non-void

The patch that introduced this warning was Maverick commit:
  commit 2aec770a575b15cb1e43c2e16dff4f00a02a8c9a
  Author: Kees Cook <kees.cook@canonical.com>
  Date:   Mon Jun 28 22:34:04 2010 -0700

    UBUNTU: SAUCE: security: create task_free security callback

Acked-by: Tim Gardner <tim.gardner@canonical.com>
Signed-off-by: Leann Ogasawara <leann.ogasawara@canonical.com>
Signed-off-by: Daniel Lezcano <daniel.lezcano@free.fr>

[ARM] implement arch_randomize_brk()

For this feature to take effect, CONFIG_COMPAT_BRK must be turned
off.  This can safely be turned off for any EABI user space versions.

Signed-off-by: Nicolas Pitre <nicolas.pitre@linaro.org>
(cherry picked from commit 990cb8acf23cab19a2930f1ed5e7dc108f89079b)

Signed-off-by: Leann Ogasawara <leann.ogasawara@canonical.com>

ARM: initial stack protector (-fstack-protector) support

This is the very basic stuff without the changing canary upon
task switch yet.  Just the Kconfig option and a constant canary
value initialized at boot time.

Signed-off-by: Nicolas Pitre <nicolas.pitre@linaro.org>
(cherry picked from commit c743f38013aeff58ef6252601e397b5ba281c633)

Signed-off-by: Leann Ogasawara <leann.ogasawara@canonical.com>

UBUNTU: SAUCE: (no-up) ARM: Resetting power_mode to its original value.

This reverts the second part of patch:

  commit 6da20c89af64b75302399369a90b9d50c1a87665
  Author: Adrian Hunter <adrian.hunter@nokia.com>
  Date:   Mon Feb 15 10:03:34 2010 -0800

    omap_hsmmc: Ensure regulator enable / disable are paired

Without this the kernel fails to initialize the SDHC card and find the
root partition.  Work is currently underway with the community to find a
real solution to the problem.  This a temporary measure to unblock
developers and will have to be reverted when the real fix gets upstream.

BugLink: https://bugs.launchpad.net/bugs/591941

Signed-off-by: Mathieu Poirier <mathieu.poirier@canonical.com>
Signed-off-by: Leann Ogasawara <leann.ogasawara@canonical.com>

UBUNTU: ubuntu: rtl8192se -- update to version 0017.0507.2010

Signed-off-by: Leann Ogasawara <leann.ogasawara@canonical.com>

UBUNTU: ubuntu: iscsitarget -- version 1.4.20.2

Update iscsitarget to the latest stable version (v1.4.20.2) from
sourceforge.

ExternalDriver: iscsi_trgt
Url: http://sourceforge.net/projects/iscsitarget/files/
Version: 1.4.20.2

Acked-by: Tim Gardner <tim.gardner@canonical.com>
Signed-off-by: Leann Ogasawara <leann.ogasawara@canonical.com>

ALSA: hda - Handle pin NID 0x1a on ALC259/269

BugLink: http://bugs.launchpad.net/bugs/582199
BugLink: http://bugs.launchpad.net/bugs/586418
BugLink: http://bugs.launchpad.net/bugs/588031

The pin NID 0x1a should be handled as well as NID 0x1b.
Also added comments.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
(cherry picked from commit b08b1637ce1c0196970348bcabf40f04b6b3d58e in
git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound-2.6.git)

Signed-off-by: David Henningsson <david.henningsson@canonical.com>
Signed-off-by: Leann Ogasawara <leann.ogasawara@canonical.com>

ALSA: hda - Handle missing NID 0x1b on ALC259 codec

BugLink: http://bugs.launchpad.net/bugs/582199
BugLink: http://bugs.launchpad.net/bugs/586418
BugLink: http://bugs.launchpad.net/bugs/588031

Since ALC259/269 use the same parser of ALC268, the pin 0x1b was ignored
as an invalid widget.  Just add this NID to handle properly.
This will add the missing mixer controls for some devices.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
(cherry picked from commit 5d4abf93ea3192cc666430225a29a4978c97c57d in
git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound-2.6.git)

Signed-off-by: David Henningsson <david.henningsson@canonical.com>
Signed-off-by: Leann Ogasawara <leann.ogasawara@canonical.com>

UBUNTU: SAUCE: blkfront: default to sd devices

EC2 defaults to "/dev/sda" rather than "/dev/xvd". Once they finally
switch then this patch should be dropped.

Signed-off-by: John Johansen <john.johansen@canonical.com>
Signed-off-by: Tim Gardner <tim.gardner@canonical.com>
Signed-off-by: Leann Ogasawara <leann.ogasawara@canonical.com>

UBUNTU: SAUCE: fix pv-ops for legacy Xen

Import fix_xen_guest_on_old_EC2.patch from fedora 14

Legacy hypervisors (RHEL 5.0 and RHEL 5.1) do not handle guest writes to
cr4 gracefully. If a guest attempts to write a bit of cr4 that is
unsupported, then the HV is so offended it crashes the domain. While
later guest kernels (such as RHEL6) don't assume the HV supports all
features, they do expect nicer responses. That assumption introduced
code that probes whether or not xsave is supported early in the boot. So
now when attempting to boot a RHEL6 guest on RHEL5.0 or RHEL5.1 an early
crash will occur.

This patch is quite obviously an undesirable hack. The real fix for this
problem should be in the HV, and is, in later HVs. However, to support
running on old HVs, RHEL6 can take this small change. No impact will
occur for running on any RHEL HV (not even RHEL 5.5 supports xsave).
There is only potential for guest performance loss on upstream Xen.

All this by way of explanation for why is this patch not going upstream.

Signed-off-by: John Johansen <john.johansen@canonical.com>
Signed-off-by: Tim Gardner <tim.gardner@canonical.com>
Signed-off-by: Leann Ogasawara <leann.ogasawara@canonical.com>

UBUNTU: SAUCE: AppArmor: Allow dfa backward compatibility with broken userspace

Allow broken Lucid userspace tools to load policy, on Maverick kernel.
The fix for http://launchpad.net/bugs/581525 blocks Lucid tools from
loading policy, this provides compatibility with Lucid tools without
reintroducing the bug.

The apparmor_parser when compiling policy could generate invalid dfas
that did not have sufficient padding to avoid invalid references, when
used by the kernel.  The kernels check to verify the next/check table
size was broken meaning invalid dfas were being created by userspace
and not caught.

To remain compatible with old tools that are not fixed, pad the loaded
dfas next/check table.  The dfa's themselves are valid except for the
high padding for potentially invalid transitions (high bounds error),
which have a maximimum is 256 entries.  So just allocate an extra null filled
256 entries for the next/check tables.  This will guarentee all bounds
are good and invalid transitions go to the null (0) state.

Signed-off-by: John Johansen <john.johansen@canonical.com>
Signed-off-by: Leann Ogasawara <leann.ogasawara@canonical.com>

UBUNTU: SAUCE: vt -- allow grub to request automatic vt_handoff

Grub may be able to select a graphics mode and paint a splash screen
for us.  If so it needs to be able to tell us it has done so.  Add
support for detecting a new graphics mode selected bit in the
screen_info passed over at boot.  Use this to automatically enable
vt_handoff mode.

Signed-off-by: Andy Whitcroft <apw@canonical.com>
Acked-by: Stefan Bader <stefan.bader@canonical.com>
Signed-off-by: Leann Ogasawara <leann.ogasawara@canonical.com>

UBUNTU: SAUCE: vt -- maintain bootloader screen mode and content until vt switch

Introduce a new VT mode KD_TRANSPARENT which endevours to leave the current
content of the framebuffer untouched.  This allows the bootloader to insert
a graphical splash and have the kernel maintain it until the OS splash
can take over.  When we finally switch away (either through programs like
plymouth or manually) the content is lost and the VT reverts to text mode.

Signed-off-by: Andy Whitcroft <apw@canonical.com>
Acked-by: Stefan Bader <stefan.bader@canonical.com>
Signed-off-by: Leann Ogasawara <leann.ogasawara@canonical.com>