This reverts commit cf450136.

It breaks reboot on at least one Thinkpad T43, as reported by Jörg Otte:
 "On reboot it shuts down as normal.
  The last lines displayed are:

  >Unmounting temporary filesystems.. [OK]
  >Deactivating swap...               [OK]
  >Unmounting local filesystems...    [OK]
  >Will now restart
  >    Restarting system

  Then I hear it accessing the cd-drive, but then it's being stuck."

Jörg bisected the regression to this commit.

That commit fixes another machine (see

  https://bugzilla.kernel.org/show_bug.cgi?id=11533

for details) that has a BIOS bug and doesn't support ACPI reset.
However, at least one of those other reporters no longer even has the
machine in question, and had a different workaround to begin with.
Besides, it clearly was a buggy BIOS.  Let's not break the correct case
to fix that case.
Reported-and-bisected-by: NJörg Otte <jrg.otte@googlemail.com>
Cc: linux-acpi@vger.kernel.org
Cc: Len Brown <lenb@kernel.org>
Cc: Peter Anvin <hpa@zytor.com>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Fixed too small hardcoded timeout values for usb_control_msg
in driver for SiliconLabs cp210x-based usb-to-serial adapters.
Replaced with USB_CTRL_GET_TIMEOUT/USB_CTRL_SET_TIMEOUT.
Signed-off-by: NYuri Matylitski <ym@tekinsoft.com>
Acked-by: NKirill A. Shutemov <kirill@shutemov.name>
Cc: stable <stable@vger.kernel.org>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

linux/drivers/xen/manage.c: In function 'do_suspend':
linux/drivers/xen/manage.c:160:5: warning: 'si.cancelled' may be used uninitialized in this function
Signed-off-by: NKonrad Rzeszutek Wilk <konrad.wilk@oracle.com>

A rather annoying and common case is when booting a PVonHVM guest
and exposing the PV KBD and PV VFB - as broken toolstacks don't
always initialize the backends correctly.

Normally The HVM guest is using the VGA driver and the emulated
keyboard for this (though upstream version of QEMU implements
PV KBD, but still uses a VGA driver). We provide a very basic
two-stage wait mechanism - where we wait for 30 seconds for all
devices, and then for 270 for all them except the two mentioned.

That allows us to wait for the essential devices, like network
or disk for the full 6 minutes.

To trigger this, put this in your guest config:

vfb = [ 'vnc=1, vnclisten=0.0.0.0 ,vncunused=1']

instead of this:
vnc=1
vnclisten="0.0.0.0"

CC: stable@kernel.org
Acked-by: NStefano Stabellini <stefano.stabellini@eu.citrix.com>
[v3: Split delay in non-essential (30 seconds) and essential
 devices per Ian and Stefano suggestion]
[v4: Added comments per Stefano suggestion]
Signed-off-by: NKonrad Rzeszutek Wilk <konrad.wilk@oracle.com>

The refactoring of the nv50 logic, introduced in 8663bc7c, modified the
test for the special lane map used on some Apple computers with Nvidia
chipsets. The tested MBA3,1 would still boot, but resume from suspend
stopped working. This patch restores the old test, which fixes the problem.
Signed-off-by: NHenrik Rydberg <rydberg@euromail.se>
Acked-by: NBen Skeggs <bskeggs@redhat.com>
Signed-off-by: NDave Airlie <airlied@redhat.com>

When the force changes went in back in 3.3.0, we ended up returning
disconnected in the !force case, and the connected in when forced,
as it hit the hardcoded check.

Fix it so all exits go via the hardcoded check and stop spurious
modesets on platforms with hardcoded EDIDs.

Reported-by: Evan McNabb (Red Hat)
Reviewed-by: NAlex Deucher <alexander.deucher@amd.com>
Cc: stable@vger.kernel.org
Signed-off-by: NDave Airlie <airlied@redhat.com>

During DRM release, all the FBs and gem objects are released. If
a gem object is being used as a FB and set to a crtc, it must not
be freed before releasing the framebuffer first.

If FBs are released first, the crtc using the FB is disabled first
so now the GEM object can be freed safely. The CRTC will be enabled
again when the driver restores fbdev mode.
Signed-off-by: NPrathyush K <prathyush.k@samsung.com>
Signed-off-by: NDave Airlie <airlied@redhat.com>

NUL-terminate after strncpy.

If the parameter "profile" has length 16 or more, then strncpy
leaves "string" with no NUL terminator, so the following search
for '\n' may read beyond the end of that 16-byte buffer.
If it finds a newline there, then it will also write beyond the
end of that stack buffer.
Signed-off-by: NJim Meyering <meyering@redhat.com>
Signed-off-by: NDave Airlie <airlied@redhat.com>

The output of "make versioncheck" points a incorrect include of
version.h in the drivers/gpu/drm/gma500/mdfld_dsi_output.h:

drivers/gpu/drm/gma500/mdfld_dsi_output.h: 32 linux/version.h not needed.

If we take a look in the file, we can agree to remove it.

Cc: David Airlie <airlied@linux.ie>
Cc: <dri-devel@lists.freedesktop.org>
Signed-off-by: NMarcos Paulo de Souza <marcos.souza.org@gmail.com>
Signed-off-by: NDave Airlie <airlied@redhat.com>

If AGP is placed in the middle, the size_af is off-by-one, it results
in VRAM being placed at 0x7fffffff instead of 0x8000000.

v2: fix the vram_start setup.
v3: also fix r7xx & newer ASIC

Reported-by: russiane39 on #radeon
Signed-off-by: NJerome Glisse <jglisse@redhat.com>
Signed-off-by: NDave Airlie <airlied@redhat.com>

Free event and restore event_space only when page_flip->flags has
DRM_MODE_PAGE_FLIP_EVENT if page_flip() is failed.
Signed-off-by: NJoonyoung Shim <jy0922.shim@samsung.com>
Signed-off-by: NKyungmin Park <kyungmin.park@samsung.com>
Reviewed-by: NMichel Dänzer <michel.daenzer@amd.com>
Signed-off-by: NDave Airlie <airlied@redhat.com>

The check of the encoder type in the commit [e00e8b5e: drm/radeon/kms:
fix analog load detection on DVI-I connectors] is obviously wrong, and
it's the culprit of the regression on my workstation with DVI-analog
connection resulting in the blank output.

Fixed the typo now.
Signed-off-by: NTakashi Iwai <tiwai@suse.de>
Reviewed-by: NAlex Deucher <alexander.deucher@amd.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: NDave Airlie <airlied@redhat.com>

Commit 4f5ca836 "HID: hid-input: add support for HID devices reporting
Battery Strength" added the CONFIG_HID_BATTERY_STRENGTH option to report
the battery strength of HID devices.  The commit log explicitly mentions
it not working properly with recent userspace, but it is default y
anyway.  This is rather odd, and actually causes problems on real
systems.

This works around Fedora bug
https://bugzilla.redhat.com/show_bug.cgi?id=806295Signed-off-by: NJosh Boyer <jwboyer@redhat.com>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

Allows this module to load correctly with certain debugging options on.

Reported on irc by scientes
Signed-off-by: NDave Airlie <airlied@redhat.com>

This call is not needed; the IRQ controller should (and does) set up
interrupts correctly. set_irq_flags() isn't exported to modules, to
this also fixes compilation of ehci-tegra.c as a module.
Signed-off-by: NStephen Warren <swarren@nvidia.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Current probing code is setting URB_NO_TRANSFER_DMA_MAP flag into a wrong urb
structure, and this causes BUG_ON with some USB host implementations.
This patch fixes the issue.
Signed-off-by: NTomoki Sekiyama <tomoki.sekiyama@gmail.com>
Cc: stable <stable@vger.kernel.org>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Removes allocation of coherent buffer for the control-request setup-packet
buffer from the yurex driver. Using coherent buffers for setup-packet is
obsolete and does not work with some USB host implementations.
Signed-off-by: NTomoki Sekiyama <tomoki.sekiyama@gmail.com>
Cc: stable <stable@vger.kernel.org>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Free the two previously allocated buffers before exiting the function in an
error case.
Signed-off-by: NJulia Lawall <Julia.Lawall@lip6.fr>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Since commit 28c56ea1
(powerpc/usb: fix bug of kernel hang when initializing usb)
the kernel crashes on mpc5121e. mpc5121e doesn't have system interface
registers, accessing this register address space cause the machine check
exception and a kernel crash:
...
[    1.294596] ehci_hcd: USB 2.0 'Enhanced' Host Controller (EHCI) Driver
[    1.316491] fsl-ehci fsl-ehci.0: Freescale On-Chip EHCI Host Controller
[    1.337334] fsl-ehci fsl-ehci.0: new USB bus registered, assigned bus number 1
[    1.358548] Machine check in kernel mode.
[    1.375917] Caused by (from SRR1=49030): Transfer error ack signal
[    1.395505] Oops: Machine check, sig: 7 [#1]
[    1.413113] MPC5121 ADS
[    1.428718] Modules linked in:
[    1.444841] NIP: c026efc4 LR: c0278b50 CTR: 00000000
[    1.463342] REGS: df837ba0 TRAP: 0200   Not tainted  (3.3.0-08839-gb5174fa3)
[    1.484083] MSR: 00049030 <EE,ME,IR,DR>  CR: 42042022  XER: 20000000
[    1.504099] TASK = df834000[1] 'swapper' THREAD: df836000
[    1.509667] GPR00: 1c000000 df837c50 df834000 df9d74e0 00000003 00000010 00000000 00000000
[    1.531650] GPR08: 00000020 00000000 c037cdd8 e1088000 22042028 1001a69c 00000000 00000000
[    1.553762] GPR16: 1ffbce70 00000000 1fef5b28 1fef3e08 00000000 00000000 1ffcbc7c c045b264
[    1.575824] GPR24: 0000008b 00000002 c04a7dd0 e1088000 df33c960 df9d74e0 00000000 df9d7400
[    1.612295] NIP [c026efc4] ehci_fsl_setup_phy+0x110/0x124
[    1.632454] LR [c0278b50] ehci_fsl_setup+0x29c/0x304
[    1.652065] Call Trace:
[    1.668923] [df837c50] [c0278a40] ehci_fsl_setup+0x18c/0x304 (unreliable)
[    1.690332] [df837c70] [c025cba4] usb_add_hcd+0x1f0/0x66c
[    1.710377] [df837cb0] [c0277ab8] ehci_fsl_drv_probe+0x180/0x308
[    1.731322] [df837ce0] [c01fc7a8] platform_drv_probe+0x20/0x30
[    1.752202] [df837cf0] [c01fb0ac] driver_probe_device+0x8c/0x214
[    1.773491] [df837d10] [c01f956c] bus_for_each_drv+0x6c/0xa8
[    1.794279] [df837d40] [c01fafdc] device_attach+0xb4/0xd8
[    1.814574] [df837d60] [c01fa44c] bus_probe_device+0xa4/0xb4
[    1.835343] [df837d80] [c01f87a8] device_add+0x52c/0x5dc
[    1.855462] [df837dd0] [c01fcd58] platform_device_add+0x124/0x1d0
[    1.876558] [df837df0] [c036dcec] fsl_usb2_device_register+0xa0/0xd4
[    1.897512] [df837e10] [c036df28] fsl_usb2_mph_dr_of_probe+0x208/0x264
[    1.918253] [df837e90] [c01fc7a8] platform_drv_probe+0x20/0x30
[    1.938300] [df837ea0] [c01fb0ac] driver_probe_device+0x8c/0x214
[    1.958511] [df837ec0] [c01fb2f0] __driver_attach+0xbc/0xc0
[    1.978088] [df837ee0] [c01f9608] bus_for_each_dev+0x60/0x9c
[    1.997589] [df837f10] [c01fab88] driver_attach+0x24/0x34
[    2.016757] [df837f20] [c01fa744] bus_add_driver+0x1ac/0x274
[    2.036339] [df837f50] [c01fb898] driver_register+0x88/0x150
[    2.056052] [df837f70] [c01fcabc] platform_driver_register+0x68/0x78
[    2.076650] [df837f80] [c0446500] fsl_usb2_mph_dr_driver_init+0x18/0x28
[    2.097734] [df837f90] [c0003988] do_one_initcall+0x148/0x1b0
[    2.117934] [df837fc0] [c042d89c] kernel_init+0xfc/0x190
[    2.137667] [df837ff0] [c000d2c4] kernel_thread+0x4c/0x68
[    2.157240] Instruction dump:
[    2.174119] 90050004 4e800020 2f840003 419e0014 2f840004 409eff64 6400c000 4bffff5c
[    2.196000] 64001000 7c0004ac 812b0500 0c090000 <4c00012c> 61290200 7c0004ac 912b0500
[    2.218100] ---[ end trace 21659aedb84ad816 ]---
[    2.237089]
[    3.232940] Kernel panic - not syncing: Attempted to kill init! exitcode=0x00000007
[    3.232954]
[    3.271575] Rebooting in 1 seconds..

Check pdata->have_sysif_regs flag before accessing system interface
registers.
Signed-off-by: NAnatolij Gustschin <agust@denx.de>
Cc: Shengzhou Liu <Shengzhou.Liu@freescale.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Fatal errors such as a device disconnect must not trigger
error handling. The error returns must be checked.
Signed-off-by: NOliver Neukum <oneukum@suse.de>
Cc: stable <stable@vger.kernel.org>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

del_timer_sync() cannot be used in interrupt.
Replace it with del_timer() and a flag
Signed-off-by: NOliver Neukum <oneukum@suse.de>
Cc: stable <stable@vger.kernel.org>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

This patch (as1544) fixes a problem affecting some EHCI controllers.
They can generate interrupts whenever the STS_FLR status bit is turned
on, even though that bit is masked out in the Interrupt Enable
register.

Since the driver doesn't use STS_FLR anyway, the patch changes the
interrupt routine to clear that bit whenever it is set, rather than
leaving it alone.
Signed-off-by: NAlan Stern <stern@rowland.harvard.edu>
Reported-and-tested-by: NTomoya MORINAGA <tomoya.rohm@gmail.com>
Cc: stable <stable@vger.kernel.org>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

drivers/block/xen-blkback/xenbus.c: In function 'xen_blkbk_discard':
drivers/block/xen-blkback/xenbus.c:419:4: warning: passing argument 1 of 'dev_warn' makes pointer from integer without a cast
+[enabled by default]
include/linux/device.h:894:5: note: expected 'const struct device *' but argument is of type 'long int'

It is unclear how that mistake made it in. It surely is wrong.
Acked-by: NJens Axboe <axboe@kernel.dk>
Reported-by: NStephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: NKonrad Rzeszutek Wilk <konrad.wilk@oracle.com>

Forbid port runtime pm by default because it has known hotplug issue.
User can allow it by, for example

echo auto > /sys/devices/pci0000:00/0000:00:1f.2/ata2/power/control
Signed-off-by: NLin Ming <ming.m.lin@intel.com>
Signed-off-by: NJeff Garzik <jgarzik@redhat.com>

The following build warning is seen in some configurations.

drivers/hwmon/ads1015.c: In function 'show_in':
drivers/hwmon/ads1015.c:129: warning: 'in' may be used uninitialized in this function

Fix by separating the register read function from the code converting the result
into mV.
Signed-off-by: NGuenter Roeck <guenter.roeck@ericsson.com>
Cc: Dirk Eibach <eibach@gdsys.de>
Reviewed-by: NRobert Coulson <robert.coulson@ericsson.com>

Newer BKDG[1] versions recommend a different initialization value for
the running average range register in the northbridge. This improves
the power reading by avoiding counter saturations resulting in bogus
values for anything below about 80% of TDP power consumption.
Updated BIOSes will have this new value set up from the beginning,
but meanwhile we correct this value ourselves.
This needs to be done on all northbridges, even on those where the
driver itself does not register at.

This fixes the driver on all current machines to provide proper
values for idle load.

[1]
http://support.amd.com/us/Processor_TechDocs/42301_15h_Mod_00h-0Fh_BKDG.pdf
Chapter 3.8: D18F5xE0 Processor TDP Running Average (p. 452)
Signed-off-by: NAndre Przywara <andre.przywara@amd.com>
Acked-by: NJean Delvare <khali@linux-fr.org>
[guenter.roeck@ericsson.com: Removed unnecessary return statement]
Signed-off-by: NGuenter Roeck <guenter.roeck@ericsson.com>
Cc: stable@vger.kernel.org # 3.0+

This reverts commit 6fe0d062.

Paul bisected this regression.

The conversion was done blindly and is wrong, as it does not provide a
primary handler to disable the level type irq on the device level.
Neither does it set the IRQF_ONESHOT flag which handles that at the irq
line level.  This can't be done as the interrupt might be shared, though
we might extend the core to force it.

So an interrupt on this line will wake up the thread, but immediately
unmask the irq after that.  Due to the interrupt being level type the
hardware interrupt is raised over and over and prevents the irq thread
from handling it.  Fail.

request_irq() unfortunately does not refuse such a request and the patch
was obviously never tested with real interrupts.
Bisected-by: NPaul Bolle <pebolle@tiscali.nl>
Signed-off-by: NThomas Gleixner <tglx@linutronix.de>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Not only do the pageflip work without it at non-native modes (i.e. with
the panel fitter enabled), it also causes normal (non-pageflipped)
modesets to fail.
Reported-by: NAdam Jackson <ajax@redhat.com>
Tested-by: NAdam Jackson <ajax@redhat.com>
Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Wanted-by-for-fixes: Dave Airlie <airlied@gmail.com>
Signed-Off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

This patch (as1542) changes the criterion ehci-hcd uses to tell when
it needs to resume the controller's root hub.  A resume is needed when
a port status change is detected, obviously, but only if the root hub
is currently suspended.

Right now the driver tests whether the root hub is running, and that
is not the correct test.  In particular, if the controller has died
then the root hub should not be restarted.  In addition, some buggy
hardware occasionally requires the root hub to be running and
sending out SOF packets even while it is nominally supposed to be
suspended.

In the end, the test needs to be changed.  Rather than checking whether
the root hub is currently running, the driver will now check whether
the root hub is currently suspended.  This will yield the correct
behavior in all cases.
Signed-off-by: NAlan Stern <stern@rowland.harvard.edu>
CC: Peter Chen <B29397@freescale.com>
Cc: stable <stable@vger.kernel.org>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

These devices have a number of non serial interfaces as well.  Use
the existing "Direct IP" blacklist to prevent binding to interfaces
which are handled by other drivers.

We also extend the "Direct IP" blacklist with with interfaces only
seen in "QMI" mode, assuming that these devices use the same
interface numbers for serial interfaces both in "Direct IP" and in
"QMI" mode.
Signed-off-by: NBjørn Mork <bjorn@mork.no>
Cc: stable <stable@vger.kernel.org>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

A large `nents' from userspace could overflow the allocation size,
leading to memory corruption.

| alloc_sglist()
| usbtest_ioctl()

Use kmalloc_array() to avoid the overflow.
Signed-off-by: NXi Wang <xi.wang@gmail.com>
Acked-by: NAlan Stern <stern@rowland.harvard.edu>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Avoid overflowing context.count = param->sglen * param->iterations,
where both `sglen' and `iterations' are from userspace.

| test_ctrl_queue()
| usbtest_ioctl()

Keep -EOPNOTSUPP for error code.
Signed-off-by: NXi Wang <xi.wang@gmail.com>
Acked-by: NAlan Stern <stern@rowland.harvard.edu>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

This patch (as154) fixes a self-deadlock that occurs when userspace
writes to the bConfigurationValue sysfs attribute for a hub with
children.  The task tries to lock the bandwidth_mutex at a time when
it already owns the lock:

	The attribute's method calls usb_set_configuration(),
	which calls usb_disable_device() with the bandwidth_mutex
	held.

	usb_disable_device() unregisters the existing interfaces,
	which causes the hub driver to be unbound.

	The hub_disconnect() routine calls hub_quiesce(), which
	calls usb_disconnect() for each of the hub's children.

	usb_disconnect() attempts to acquire the bandwidth_mutex
	around a call to usb_disable_device().

The solution is to make usb_disable_device() acquire the mutex for
itself instead of requiring the caller to hold it.  Then the mutex can
cover only the bandwidth deallocation operation and not the region
where the interfaces are unregistered.

This has the potential to change system behavior slightly when a
config change races with another config or altsetting change.  Some of
the bandwidth released from the old config might get claimed by the
other config or altsetting, make it impossible to restore the old
config in case of a failure.  But since we don't try to recover from
config-change failures anyway, this doesn't matter.

[This should be marked for stable kernels that contain the commit
fccf4e86 "USB: Free bandwidth when
usb_disable_device is called."
That commit was marked for stable kernels as old as 2.6.32.]
Signed-off-by: NAlan Stern <stern@rowland.harvard.edu>
Signed-off-by: NSarah Sharp <sarah.a.sharp@linux.intel.com>
Cc: stable <stable@vger.kernel.org>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Since we are using the m2p_override we do have struct pages
corresponding to the user vma mmap'ed by gntdev.

Removing the VM_PFNMAP flag makes get_user_pages work on that vma.
An example test case would be using a Xen userspace block backend
(QDISK) on a file on NFS using O_DIRECT.

CC: stable@kernel.org
Signed-off-by: NStefano Stabellini <stefano.stabellini@eu.citrix.com>
Signed-off-by: NKonrad Rzeszutek Wilk <konrad.wilk@oracle.com>

Jump to the label ini_nomem as done on the failure of the page allocations
above.

The code at ini_nomem is modified to accommodate different return values.
Signed-off-by: NJulia Lawall <Julia.Lawall@lip6.fr>
Signed-off-by: NKonrad Rzeszutek Wilk <konrad.wilk@oracle.com>

Some shortcomings introduced into pci_restore_state() by commit
26f41062 ("PCI: check for pci bar restore completion and retry")
have been fixed by recent commit ebfc5b80 ("PCI: Fix regression in
pci_restore_state(), v3"), but that commit treats all PCI devices as
those with Type 0 configuration headers.

That is not entirely correct, because Type 1 and Type 2 headers have
different layouts.  In particular, the area occupied by BARs in Type 0
config headers contains the secondary status register in Type 1 ones and
it doesn't make sense to retry the restoration of that register even if
the value read back from it after a write is not the same as the written
one (it very well may be different).

For this reason, make pci_restore_state() only retry the restoration
of BARs for Type 0 config headers.  This effectively makes it behave
as before commit 26f41062 for all header types except for Type 0.
Tested-by: NMikko Vinni <mmvinni@yahoo.com>
Signed-off-by: NRafael J. Wysocki <rjw@sisk.pl>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Upon lastclose(), we switch back to the fbcon configuration. This
requires taking the mode_config lock in order to serialise the change
with output probing elsewhere.
Reported-by: NOleksij Rempel <bug-track@fisher-privat.net>
References: https://bugs.freedesktop.org/show_bug.cgi?id=48652Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Cc: stable@kernel.org
Acked-by: NJesse Barnes <jbarnes@virtuousgeek.org>
Signed-Off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

Forget to unreserve after pinning.  This can lead to problems in
soft reset and resume.

v2: rework patch as per Michel's suggestion.
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>
Reviewed-by: NMichel Dänzer <michel.daenzer@amd.com>
Signed-off-by: NDave Airlie <airlied@redhat.com>

My rv515 card is very flaky with msi enabled. Every so often it loses a rearm
and never comes back, manually banging the rearm brings it back.
Reviewed-by: NAlex Deucher <alexander.deucher@amd.com>
Cc: stable@vger.kernel.org
Signed-off-by: NDave Airlie <airlied@redhat.com>

This regression has been introduced in

commit ca9bfa7e
Author: Daniel Vetter <daniel.vetter@ffwll.ch>
Date:   Sat Jan 28 14:49:20 2012 +0100

    drm/i915: fixup interlaced vertical timings confusion, part 1

Unfortunately that commit failed to take into account that the lvds
code does some special adjustements to the crtc timings for upscaling
an centering.

Fix this by explicitly computing crtc timings in the lvds mode fixup
function and setting a special flag in mode->private_flags if the crtc
timings have been adjusted.

v2: Add a comment to explain the new mode driver private flag,
suggested by Eugeni Dodonov.

v3: Kill the confusing and now redundant set_crtcinfo call in
intel_fixed_panel_mode, noticed by Chris Wilson.
Reported-and-Tested-by: NHans de Bruin <jmdebruin@xmsnet.nl>
Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=43071Reviewed-by: NEugeni Dodonov <eugeni.dodonov@intel.com>
Reviewed-by: NChris Wilson <chris@chris-wilson.co.uk>
Signed-Off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>