The vast majority of platforms are not supplying ACPI _PXM (proximity)
information corresponding to host bridge (PNP0A03/PNP0A08) devices
resulting in sysfs "numa_node" values of -1 (NUMA_NO_NODE):

  # for i in /sys/devices/pci0000\:00/*/numa_node; do cat $i; done | uniq
  -1

  # find /sys/ -name "numa_node" | while read fname; do cat $fname; \
    done | uniq
  -1

AMD based platforms provide a fall-back for this situation via amd_bus.c.
These platforms snoop out the information by directly reading specific
registers from the Northbridge and caching them via alloc_pci_root_info().

Later during boot processing when host bridges are discovered -
pci_acpi_scan_root() - the kernel looks for their corresponding ACPI _PXM
method - drivers/acpi/numa.c::acpi_get_node().  If the BIOS supplied a _PXM
method then that node (proximity) value is associated.  If the BIOS did not
supply a _PXM method *and* the platform is AMD-based, the fall-back cached
values obtained directly from the Northbridge are used; otherwise,
"NUMA_NO_NODE" is associated.

There are a number of issues with this fall-back mechanism the most notable
being that amd_bus.c extracts a 3-bit number from a CPU register and uses
it as the node number.  The node numbers used by Linux are logical and
there's no reason they need to be identical to settings in the CPU
registers.  So if we have some node information obtained in the normal way
(from _PXM, SLIT, SRAT, etc.) and some from amd_bus.c, there's no reason to
believe they will be compatible.

This patch warns when this situation occurs:

  pci_root PNP0A08:00: [Firmware Bug]: no _PXM; falling back to node 0 from hardware (may be inconsistent with ACPI node numbers)

Link: https://bugzilla.kernel.org/show_bug.cgi?id=72051Signed-off-by: NMyron Stowe <myron.stowe@redhat.com>
Signed-off-by: NSuravee Suthikulpanit <Suravee.Suthikulpanit@amd.com>
Signed-off-by: NBjorn Helgaas <bhelgaas@google.com>

Consistently put the function type, name, and parameters on one line,
wrapping only as necessary.
Signed-off-by: NBjorn Helgaas <bhelgaas@google.com>

Reword comments so they make sense.
Signed-off-by: NBjorn Helgaas <bhelgaas@google.com>

Signed-off-by: NBjorn Helgaas <bhelgaas@google.com>

The PCI host bridge code doesn't care about _PXM values directly; it only
needs to know what NUMA node the hardware is on.

This uses acpi_get_node() directly and removes the _PXM stuff.
Signed-off-by: NBjorn Helgaas <bhelgaas@google.com>
Acked-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>

NUMA_NO_NODE is the usual value for "we don't know what node this is on,"
e.g., it is the error return from acpi_get_node().  This changes uses of -1
to NUMA_NO_NODE.  NUMA_NO_NODE is #defined to be -1 already, so this is not
a functional change.
Signed-off-by: NBjorn Helgaas <bhelgaas@google.com>
Acked-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>

There are no callers of get_mp_bus_to_node(), so we no longer need
mp_bus_to_node[], get_mp_bus_to_node(), or set_mp_bus_to_node().
This removes them.
Signed-off-by: NBjorn Helgaas <bhelgaas@google.com>
Acked-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>

This replaces all uses of get_mp_bus_to_node() with x86_pci_root_bus_node().

I think these uses are all on root buses, except possibly for blind
probing, where NUMA node information is unimportant.
Signed-off-by: NBjorn Helgaas <bhelgaas@google.com>
Acked-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>

Modify struct acpi_dev_node to contain a pointer to struct acpi_device
associated with the given device object (that is, its ACPI companion
device) instead of an ACPI handle corresponding to it.  Introduce two
new macros for manipulating that pointer in a CONFIG_ACPI-safe way,
ACPI_COMPANION() and ACPI_COMPANION_SET(), and rework the
ACPI_HANDLE() macro to take the above changes into account.
Drop the ACPI_HANDLE_SET() macro entirely and rework its users to
use ACPI_COMPANION_SET() instead.  For some of them who used to
pass the result of acpi_get_child() directly to ACPI_HANDLE_SET()
introduce a helper routine acpi_preset_companion() doing an
equivalent thing.

The main motivation for doing this is that there are things
represented by struct acpi_device objects that don't have valid
ACPI handles (so called fixed ACPI hardware features, such as
power and sleep buttons) and we would like to create platform
device objects for them and "glue" them to their ACPI companions
in the usual way (which currently is impossible due to the
lack of valid ACPI handles).  However, there are more reasons
why it may be useful.

First, struct acpi_device pointers allow of much better type checking
than void pointers which are ACPI handles, so it should be more
difficult to write buggy code using modified struct acpi_dev_node
and the new macros.  Second, the change should help to reduce (over
time) the number of places in which the result of ACPI_HANDLE() is
passed to acpi_bus_get_device() in order to obtain a pointer to the
struct acpi_device associated with the given "physical" device,
because now that pointer is returned by ACPI_COMPANION() directly.
Finally, the change should make it easier to write generic code that
will build both for CONFIG_ACPI set and unset without adding explicit
compiler directives to it.
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>
Acked-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Tested-by: Mika Westerberg <mika.westerberg@linux.intel.com> # on Haswell
Reviewed-by: NMika Westerberg <mika.westerberg@linux.intel.com>
Reviewed-by: Aaron Lu <aaron.lu@intel.com> # for ATA and SDIO part

Previously we coalesced windows by expanding the first overlapping one and
making the second invalid.  But we never look at the expanded first window
again, so we fail to notice other windows that overlap it.  For example, we
coalesced these:

  [io  0x0000-0x03af] // #0
  [io  0x03e0-0x0cf7] // #1
  [io  0x0000-0xdfff] // #2

into these, which still overlap:

  [io  0x0000-0xdfff] // #0
  [io  0x03e0-0x0cf7] // #1

The fix is to expand the *second* overlapping resource and ignore the
first, so we get this instead with no overlaps:

  [io  0x0000-0xdfff] // #2

[bhelgaas: changelog]
Reference: https://bugzilla.kernel.org/show_bug.cgi?id=62511Signed-off-by: NAlexey Neyman <stilor@att.net>
Signed-off-by: NBjorn Helgaas <bhelgaas@google.com>

Based on a patch by Jon Mason (see URL below).

All users of pcie_bus_configure_settings() pass arguments of the form
"bus, bus->self->pcie_mpss".  The "mpss" argument is redundant since we
can easily look it up internally.  In addition, all callers check
"bus->self" for NULL, which we can also do internally.

This patch simplifies the interface and the callers.  No functional change.

Reference: http://lkml.kernel.org/r/1317048850-30728-2-git-send-email-mason@myri.comSigned-off-by: NBjorn Helgaas <bhelgaas@google.com>

We should increase info->res_num before we checking pci_use_crs return
when pci=nocrs set.

No functional change, since we don't use res_num and res_offset[]
in the "!pci_use_crs" case anyway, but this makes the code read better.
Signed-off-by: NYijing Wang <wangyijing@huawei.com>
Signed-off-by: NBjorn Helgaas <bhelgaas@google.com>
Cc: Yinghai Lu <yinghai@kernel.org>
Cc: Jiang Liu <liuj97@gmail.com>
Cc: "Rafael J. Wysocki" <rafael.j.wysocki@intel.com>
Cc: Feng Tang <feng.tang@intel.com>

The ACPI handles of PCI root bridges need to be known to
acpi_bind_one(), so that it can create the appropriate
"firmware_node" and "physical_node" files for them, but currently
the way it gets to know those handles is not exactly straightforward
(to put it lightly).

This is how it works, roughly:

  1. acpi_bus_scan() finds the handle of a PCI root bridge,
     creates a struct acpi_device object for it and passes that
     object to acpi_pci_root_add().

  2. acpi_pci_root_add() creates a struct acpi_pci_root object,
     populates its "device" field with its argument's address
     (device->handle is the ACPI handle found in step 1).

  3. The struct acpi_pci_root object created in step 2 is passed
     to pci_acpi_scan_root() and used to get resources that are
     passed to pci_create_root_bus().

  4. pci_create_root_bus() creates a struct pci_host_bridge object
     and passes its "dev" member to device_register().

  5. platform_notify(), which for systems with ACPI is set to
     acpi_platform_notify(), is called.

So far, so good.  Now it starts to be "interesting".

  6. acpi_find_bridge_device() is used to find the ACPI handle of
     the given device (which is the PCI root bridge) and executes
     acpi_pci_find_root_bridge(), among other things, for the
     given device object.

  7. acpi_pci_find_root_bridge() uses the name (sic!) of the given
     device object to extract the segment and bus numbers of the PCI
     root bridge and passes them to acpi_get_pci_rootbridge_handle().

  8. acpi_get_pci_rootbridge_handle() browses the list of ACPI PCI
     root bridges and finds the one that matches the given segment
     and bus numbers.  Its handle is then used to initialize the
     ACPI handle of the PCI root bridge's device object by
     acpi_bind_one().  However, this is *exactly* the ACPI handle we
     started with in step 1.

Needless to say, this is quite embarassing, but it may be avoided
thanks to commit f3fd0c8a (ACPI: Allow ACPI handles of devices to be
initialized in advance), which makes it possible to initialize the
ACPI handle of a device before passing it to device_register().

Accordingly, add a new __weak routine, pcibios_root_bridge_prepare(),
defaulting to an empty implementation that can be replaced by the
interested architecutres (x86 and ia64 at the moment) with functions
that will set the root bridge's ACPI handle before its dev member is
passed to device_register().  Make both x86 and ia64 provide such
implementations of pcibios_root_bridge_prepare() and remove
acpi_pci_find_root_bridge() and acpi_get_pci_rootbridge_handle() that
aren't necessary any more.

Included is a fix for breakage on systems with non-ACPI PCI host
bridges from Bjorn Helgaas.
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>
Signed-off-by: NBjorn Helgaas <bhelgaas@google.com>

CONFIG_HOTPLUG is going away as an option.  As a result, the __dev*
markings need to be removed.

This change removes the use of __devinit, __devexit_p, __devinitconst,
and __devexit from these drivers.

Based on patches originally written by Bill Pemberton, but redone by me
in order to handle some of the coding style issues better, by hand.

Cc: Bill Pemberton <wfp5p@virginia.edu>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Daniel Drake <dsd@laptop.org>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

This reverts commit 0a290ac4 on
the basis of the following comment from Bjorn Helgaas:

  Here's my reasoning: this is a CheckPoint product, and it looks
  like an appliance, not really a general-purpose machine.  The issue
  has apparently been there from day one, and the kernel shipped on
  the machine complains noisily about the issue, but apparently
  nobody bothered to investigate it.

  This corruption will clearly break other ACPI-related things.  We
  can sort of work around this one (though the workaround does
  prevent us from doing any PCI resource reassignment), but we have
  no idea what the other lurking ACPI issues are (and we have no
  assurance that *only* ACPI things are broken -- maybe the
  memory corruption affects other unknown things).  It may take
  significant debugging effort to identify the next problem.

  The only report I've seen (this one) is apparently from a
  CheckPoint employee, so it's not clear that anybody else is trying
  to run upstream Linux on it.  Being a CheckPoint employee, [...]
  is probably in a position to get the BIOS fixed.

  You might still be able to convince me, but it seems like the
  benefit to a quirk for this platform is small, and it does cost
  everybody else something in code size and complexity.

References: https://bugzilla.kernel.org/show_bug.cgi?id=47981#c36Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>

This is to fix a regression https://bugzilla.kernel.org/show_bug.cgi?id=47981

The CheckPoint P-20-00 works ok before new machines (2008 and later) are
forced to use the bridge _CRS info by default in 2.6.34. Add this quirk
to restore its old way of working: not using bridge _CRS info.
Signed-off-by: NFeng Tang <feng.tang@intel.com>
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>

The memory range descriptors in the _CRS control method contain an address
translation offset for host bridges.  This value is used to translate
addresses across the bridge.  The support to use _TRA values is present for
other architectures but not for X86 platforms.

For existing X86 platforms the _TRA value is zero.  Non-zero _TRA values
are expected on future X86 platforms.  This change will register that value
with the resource.
Signed-off-by: NMike Yoknis <mike.yoknis@hp.com>
Signed-off-by: NBjorn Helgaas <bhelgaas@google.com>

The xw9300 BIOS supplies _SEG methods that are incorrect, which results
in some LSI SCSI devices not being discovered.  This adds a quirk to
ignore _SEG on this machine and default to zero.

The xw9300 has three host bridges:

    ACPI: PCI Root Bridge [PCI0] (domain 0000 [bus 00-3f])
    ACPI: PCI Root Bridge [PCI1] (domain 0001 [bus 40-7f])
    ACPI: PCI Root Bridge [PCI2] (domain 0002 [bus 80-ff])

When the BIOS "ACPI Bus Segmentation" option is enabled (as it is by
default), the _SEG methods of the PCI1 and PCI2 bridges return 1 and 2,
respectively.  However, the BIOS implementation appears to be incomplete,
and we can't enumerate devices in those domains.

But if we assume PCI1 and PCI2 really lead to buses in domain 0,
everything works fine.  Windows XP and Vista also seem to ignore
these _SEG methods.

Reference: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=543308
Reference: https://bugzilla.kernel.org/show_bug.cgi?id=15362Reported-and-Tested-by: NSean M. Pappalardo <pegasus@renegadetech.com>
Signed-off-by: NBjorn Helgaas <bhelgaas@google.com>

Use kzalloc() so the struct resource doesn't contain garbage in
fields we don't initialize.

[bhelgaas: changelog]
Signed-off-by: NYinghai Lu <yinghai@kernel.org>
Signed-off-by: NBjorn Helgaas <bhelgaas@google.com>
Cc: x86@kernel.org

For each resource of a PCI host bridge, the arch code and PCI
code log following messages.  We don't need both, so drop the
arch-specific printing.

    pci_root PNP0A08:00: host bridge window [io 0x0000-0x03af]
    pci_bus 0000:00: root bus resource [io  0x0000-0x03af]
Reviewed-by: NYinghai Lu <yinghai@kernel.org>
Signed-off-by: NJiang Liu <liuj97@gmail.com>
Signed-off-by: NBjorn Helgaas <bhelgaas@google.com>

This patch enhances x86 arch-specific code to update MMCONFIG information
when PCI host bridge hotplug event happens.
Reviewed-by: NYinghai Lu <yinghai@kernel.org>
Signed-off-by: NJiang Liu <liuj97@gmail.com>
Signed-off-by: NBjorn Helgaas <bhelgaas@google.com>

Add the host bridge bus number aperture from _CRS to the resource list.
Like the MMIO and I/O port apertures, this will be used when assigning
resources to hot-added devices or in the case of conflicts.

Note that we always use the _CRS bus number aperture, even if we're
ignoring _CRS otherwise.

[bhelgaas: changelog]
Signed-off-by: NYinghai Lu <yinghai@kernel.org>
Signed-off-by: NBjorn Helgaas <bhelgaas@google.com>

Replace the struct pci_bus secondary/subordinate members with the
struct resource busn_res.  Later we'll build a resource tree of these
bus numbers.

[bhelgaas: changelog]
Signed-off-by: NYinghai Lu <yinghai@kernel.org>
Signed-off-by: NBjorn Helgaas <bhelgaas@google.com>

Add resource_overlaps(), which returns true if two resources overlap at all.

Use this to replace the complicated check in coalesce_windows().
Signed-Off-By: NWei Yang <weiyang@linux.vnet.ibm.com>
Signed-off-by: NBjorn Helgaas <bhelgaas@google.com>

Embed the x86 struct pci_sysdata in the struct pci_root_info so it
will be automatically freed in the remove path.
Signed-off-by: NYinghai Lu <yinghai@kernel.org>
Signed-off-by: NBjorn Helgaas <bhelgaas@google.com>

We now keep the pci_root_info struct for the entire lifetime of the
host bridge, so just embed the name in the struct rather than
allocating it separately.
Signed-off-by: NYinghai Lu <yinghai@kernel.org>
Signed-off-by: NBjorn Helgaas <bhelgaas@google.com>

1. Allocate pci_root_info instead of using stack.  We need to pass around
   info for release function.
2. Add release_pci_root_info
3. Set x86 host bridge release function to make sure root bridge
   related resources get freed during root bus removal.
Signed-off-by: NYinghai Lu <yinghai@kernel.org>
Signed-off-by: NBjorn Helgaas <bhelgaas@google.com>

Rename get_current_resources() to probe_pci_root_info.

1. Remove resource list head from pci_root_info
2. Make get_current_resources() not pass resources
3. Rename get_current_resources() to probe_pci_root_info()
Signed-off-by: NYinghai Lu <yinghai@kernel.org>
Signed-off-by: NBjorn Helgaas <bhelgaas@google.com>

In pci_scan_acpi_root(), when pci_use_crs is set, get_current_resources()
is used to get pci_root_info, and it will allocate name and resource array.

Later if pci_create_root_bus() can not create bus (could be already
there...) it will only free bus res list, but the name and res array is not
freed.

Let get_current_resource() take info pointer instead of using local info.
Signed-off-by: NYinghai Lu <yinghai@kernel.org>
Signed-off-by: NBjorn Helgaas <bhelgaas@google.com>

Carlos was getting

	WARNING: at drivers/pci/pci.c:118 pci_ioremap_bar+0x24/0x52()

when probing his sound card, and sound did not work.  After adding
pci=use_crs to the kernel command line, no more trouble.

Ok, we can add a quirk.  dmidecode output reveals that this is an MSI
MS-7253, for which we already have a quirk, but the short-sighted
author tied the quirk to a single BIOS version, making it not kick in
on Carlos's machine with BIOS V1.2.  If a later BIOS update makes it
no longer necessary to look at the _CRS info it will still be
harmless, so let's stop trying to guess which versions have and don't
have accurate _CRS tables.

Addresses https://bugtrack.alsa-project.org/alsa-bug/view.php?id=5533
Also see <https://bugzilla.kernel.org/show_bug.cgi?id=42619>.
Reported-by: NCarlos Luna <caralu74@gmail.com>
Reviewed-by: NBjorn Helgaas <bhelgaas@google.com>
Signed-off-by: NJonathan Nieder <jrnieder@gmail.com>
Signed-off-by: NJesse Barnes <jbarnes@virtuousgeek.org>

In the spirit of commit 29cf7a30 ("x86/PCI: use host bridge _CRS
info on ASUS M2V-MX SE"), this DMI quirk turns on "pci_use_crs" by
default on a board that needs it.

This fixes boot failures and oopses introduced in 3e3da00c
("x86/pci: AMD one chain system to use pci read out res").  The quirk
is quite targetted (to a specific board and BIOS version) for two
reasons:

 (1) to emphasize that this method of tackling the problem one quirk
     at a time is a little insane

 (2) to give BIOS vendors an opportunity to use simpler tables and
     allow us to return to generic behavior (whatever that happens to
     be) with a later BIOS update

In other words, I am not at all happy with having quirks like this.
But it is even worse for the kernel not to work out of the box on
these machines, so...

Reference: https://bugzilla.kernel.org/show_bug.cgi?id=42619Reported-by: NSvante Signell <svante.signell@telia.com>
Signed-off-by: NJonathan Nieder <jrnieder@gmail.com>
Signed-off-by: NBjorn Helgaas <bhelgaas@google.com>
Signed-off-by: NJesse Barnes <jbarnes@virtuousgeek.org>

warning:
unreferenced object 0xffff8801f6914200 (size 512):
  comm "swapper/0", pid 1, jiffies 4294893643 (age 2664.644s)
  hex dump (first 32 bytes):
    00 00 c0 fe 00 00 00 00 ff ff ff ff 00 00 00 00  ................
    60 58 2f f6 03 88 ff ff 00 02 00 00 00 00 00 00  `X/.............
  backtrace:
    [<ffffffff81c2408c>] kmemleak_alloc+0x26/0x43
    [<ffffffff8113764f>] __kmalloc+0x121/0x183
    [<ffffffff81ca8d93>] get_current_resources+0x5a/0xc6
    [<ffffffff81c5bedd>] pci_acpi_scan_root+0x13c/0x21c
    [<ffffffff81c2a745>] acpi_pci_root_add+0x1e1/0x421
    [<ffffffff81408f50>] acpi_device_probe+0x50/0x190
    [<ffffffff8149edc7>] really_probe+0x99/0x126
    [<ffffffff8149ef83>] driver_probe_device+0x3b/0x56
    [<ffffffff8149effd>] __driver_attach+0x5f/0x82
    [<ffffffff8149d860>] bus_for_each_dev+0x5c/0x88
    [<ffffffff8149eb87>] driver_attach+0x1e/0x20
    [<ffffffff8149e7cc>] bus_add_driver+0xca/0x21d
    [<ffffffff8149f47b>] driver_register+0x91/0xfe
    [<ffffffff81409d09>] acpi_bus_register_driver+0x43/0x45
    [<ffffffff8278bdc9>] acpi_pci_root_init+0x20/0x28
    [<ffffffff810001e7>] do_one_initcall+0x57/0x134

The system has _CRS for root buses, but they are not used because the machine
date is before the cutoff date for _CRS usage.

Try to free those unused resource arrays and names.
Reviewed-by: NBjorn Helgaas <bhelgaas@google.com>
Signed-off-by: NYinghai Lu <yinghai@kernel.org>
Signed-off-by: NJesse Barnes <jbarnes@virtuousgeek.org>

Host bridges that lead to things like the Uncore need not have any
I/O port or MMIO apertures.  For example, in this case:

    ACPI: PCI Root Bridge [UNC1] (domain 0000 [bus ff])
    PCI: root bus ff: using default resources
    PCI host bridge to bus 0000:ff
    pci_bus 0000:ff: root bus resource [io  0x0000-0xffff]
    pci_bus 0000:ff: root bus resource [mem 0x00000000-0x3fffffffffff]

we should not pretend those default resources are available on bus ff.

CC: Yinghai Lu <yinghai@kernel.org>
Signed-off-by: NBjorn Helgaas <bhelgaas@google.com>
Signed-off-by: NJesse Barnes <jbarnes@virtuousgeek.org>

x86 has two kinds of PCI root bus scanning:

(1) ACPI-based, using _CRS resources.  This used pci_create_bus(), not
    pci_scan_bus(), because ACPI hotplug needed to split the
    pci_bus_add_devices() into a separate host bridge .start() method.

    This patch parses the _CRS resources earlier, so we can build a list of
    resources and pass it to pci_create_root_bus().

    Note that as before, we parse the _CRS even if we aren't going to use
    it so we can print it for debugging purposes.

(2) All other, which used either default resources (ioport_resource and
    iomem_resource) or information read from the hardware via amd_bus.c or
    similar.  This used pci_scan_bus().

    This patch converts x86_pci_root_bus_res_quirks() (previously called
    from pcibios_fixup_bus()) to x86_pci_root_bus_resources(), which builds
    a list of resources before we call pci_scan_root_bus().

    We also use x86_pci_root_bus_resources() if we have ACPI but are
    ignoring _CRS.

CC: Yinghai Lu <yinghai.lu@oracle.com>
Signed-off-by: NBjorn Helgaas <bhelgaas@google.com>
Signed-off-by: NJesse Barnes <jbarnes@virtuousgeek.org>

This assures that a _CRS reserved host bridge window or window region is
not used if it is not addressable by the CPU.  The new code either trims
the window to exclude the non-addressable portion or totally ignores the
window if the entire window is non-addressable.

The current code has been shown to be problematic with 32-bit non-PAE
kernels on systems where _CRS reserves resources above 4GB.
Signed-off-by: NGary Hade <garyhade@us.ibm.com>
Reviewed-by: NBjorn Helgaas <bhelgaas@google.com>
Cc: Thomas Renninger <trenn@novell.com>
Cc: linux-kernel@vger.kernel.org
Cc: stable@kernel.org
Signed-off-by: NJesse Barnes <jbarnes@virtuousgeek.org>

Enabling CRS by default breaks suspend on the Thinkpad SL510.
Details in https://bugzilla.redhat.com/show_bug.cgi?id=769657Reported-by: NStefan Kirrmann <stefan.kirrmann@gmail.com>
Signed-off-by: NDave Jones <davej@redhat.com>
Signed-off-by: NJesse Barnes <jbarnes@virtuousgeek.org>

The Dell Studio 1557 also doesn't suspend correctly when CRS is enabled.
Details at https://bugzilla.redhat.com/show_bug.cgi?id=769657Reported-by: NGregory S. Hoerner <ghoerner@transcendingthought.com>
Signed-off-by: NDave Jones <davej@redhat.com>
Signed-off-by: NJesse Barnes <jbarnes@virtuousgeek.org>

Some machines don't boot unless passed pci=nocrs.
(See https://bugzilla.redhat.com/show_bug.cgi?id=770308 for details of
  one report. Waiting on dmidecode output for others).

Currently there is a DMI whitelist, even though the default is on.

v2: drop the 1536 blacklist entry, superceded by the PNP/MMCONFIG changes from
    Bjorn
Acked-by: NBjorn Helgaas <bhelgaas@google.com>
Acked-by: NIngo Molnar <mingo@elte.hu>
Signed-off-by: NDave Jones <davej@redhat.com>
Signed-off-by: NJesse Barnes <jbarnes@virtuousgeek.org>

In summary, this DMI quirk uses the _CRS info by default for the ASUS
M2V-MX SE by turning on `pci=use_crs` and is similar to the quirk
added by commit 2491762c ("x86/PCI: use host bridge _CRS info on
ASRock ALiveSATA2-GLAN") whose commit message should be read for further
information.

Since commit 3e3da00c ("x86/pci: AMD one chain system to use pci
read out res") Linux gives the following oops:

    parport0: PC-style at 0x378, irq 7 [PCSPP,TRISTATE]
    HDA Intel 0000:20:01.0: PCI INT A -> GSI 17 (level, low) -> IRQ 17
    HDA Intel 0000:20:01.0: setting latency timer to 64
    BUG: unable to handle kernel paging request at ffffc90011c08000
    IP: [<ffffffffa0578402>] azx_probe+0x3ad/0x86b [snd_hda_intel]
    PGD 13781a067 PUD 13781b067 PMD 1300ba067 PTE 800000fd00000173
    Oops: 0009 [#1] SMP
    last sysfs file: /sys/module/snd_pcm/initstate
    CPU 0
    Modules linked in: snd_hda_intel(+) snd_hda_codec snd_hwdep snd_pcm_oss snd_mixer_oss snd_pcm snd_seq_midi snd_rawmidi snd_seq_midi_event tpm_tis tpm snd_seq tpm_bios psmouse parport_pc snd_timer snd_seq_device parport processor evdev snd i2c_viapro thermal_sys amd64_edac_mod k8temp i2c_core soundcore shpchp pcspkr serio_raw asus_atk0110 pci_hotplug edac_core button snd_page_alloc edac_mce_amd ext3 jbd mbcache sha256_generic cryptd aes_x86_64 aes_generic cbc dm_crypt dm_mod raid1 md_mod usbhid hid sg sd_mod crc_t10dif sr_mod cdrom ata_generic uhci_hcd sata_via pata_via libata ehci_hcd usbcore scsi_mod via_rhine mii nls_base [last unloaded: scsi_wait_scan]
    Pid: 1153, comm: work_for_cpu Not tainted 2.6.37-1-amd64 #1 M2V-MX SE/System Product Name
    RIP: 0010:[<ffffffffa0578402>]  [<ffffffffa0578402>] azx_probe+0x3ad/0x86b [snd_hda_intel]
    RSP: 0018:ffff88013153fe50  EFLAGS: 00010286
    RAX: ffffc90011c08000 RBX: ffff88013029ec00 RCX: 0000000000000006
    RDX: 0000000000000000 RSI: 0000000000000246 RDI: 0000000000000246
    RBP: ffff88013341d000 R08: 0000000000000000 R09: 0000000000000040
    R10: 0000000000000286 R11: 0000000000003731 R12: ffff88013029c400
    R13: 0000000000000000 R14: 0000000000000000 R15: ffff88013341d090
    FS:  0000000000000000(0000) GS:ffff8800bfc00000(0000) knlGS:00000000f7610ab0
    CS:  0010 DS: 0000 ES: 0000 CR0: 000000008005003b
    CR2: ffffc90011c08000 CR3: 0000000132f57000 CR4: 00000000000006f0
    DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
    DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400
    Process work_for_cpu (pid: 1153, threadinfo ffff88013153e000, task ffff8801303c86c0)
    Stack:
     0000000000000005 ffffffff8123ad65 00000000000136c0 ffff88013029c400
     ffff8801303c8998 ffff88013341d000 ffff88013341d090 ffff8801322d9dc8
     ffff88013341d208 0000000000000000 0000000000000000 ffffffff811ad232
    Call Trace:
     [<ffffffff8123ad65>] ? __pm_runtime_set_status+0x162/0x186
     [<ffffffff811ad232>] ? local_pci_probe+0x49/0x92
     [<ffffffff8105afc5>] ? do_work_for_cpu+0x0/0x1b
     [<ffffffff8105afc5>] ? do_work_for_cpu+0x0/0x1b
     [<ffffffff8105afd0>] ? do_work_for_cpu+0xb/0x1b
     [<ffffffff8105fd3f>] ? kthread+0x7a/0x82
     [<ffffffff8100a824>] ? kernel_thread_helper+0x4/0x10
     [<ffffffff8105fcc5>] ? kthread+0x0/0x82
     [<ffffffff8100a820>] ? kernel_thread_helper+0x0/0x10
    Code: f4 01 00 00 ef 31 f6 48 89 df e8 29 dd ff ff 85 c0 0f 88 2b 03 00 00 48 89 ef e8 b4 39 c3 e0 8b 7b 40 e8 fc 9d b1 e0 48 8b 43 38 <66> 8b 10 66 89 14 24 8b 43 14 83 e8 03 83 f8 01 77 32 31 d2 be
    RIP  [<ffffffffa0578402>] azx_probe+0x3ad/0x86b [snd_hda_intel]
     RSP <ffff88013153fe50>
    CR2: ffffc90011c08000
    ---[ end trace 8d1f3ebc136437fd ]---

Trusting the ACPI _CRS information (`pci=use_crs`) fixes this problem.

    $ dmesg | grep -i crs # with the quirk
    PCI: Using host bridge windows from ACPI; if necessary, use "pci=nocrs" and report a bug

The match has to be against the DMI board entries though since the vendor entries are not populated.

    DMI: System manufacturer System Product Name/M2V-MX SE, BIOS 0304    10/30/2007

This quirk should be removed when `pci=use_crs` is enabled for machines
from 2006 or earlier or some other solution is implemented.

Using coreboot [1] with this board the problem does not exist but this
quirk also does not affect it either. To be safe though the check is
tightened to only take effect when the BIOS from American Megatrends is
used.

        15:13 < ruik> but coreboot does not need that
        15:13 < ruik> because i have there only one root bus
        15:13 < ruik> the audio is behind a bridge

        $ sudo dmidecode
        BIOS Information
                Vendor: American Megatrends Inc.
                Version: 0304
                Release Date: 10/30/2007

[1] http://www.coreboot.org/

Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=30552

Cc: stable@kernel.org (2.6.34)
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: x86@kernel.org
Signed-off-by: NPaul Menzel <paulepanter@users.sourceforge.net>
Signed-off-by: NBjorn Helgaas <bhelgaas@google.com>
Acked-by: NJesse Barnes <jbarnes@virtuousgeek.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Commit b03e7495 ("PCI: Set PCI-E Max Payload Size on fabric")
introduced a potential NULL pointer dereference in calls to
pcie_bus_configure_settings due to attempts to access pci_bus self
variables when the self pointer is NULL.

To correct this, verify that the self pointer in pci_bus is non-NULL
before dereferencing it.
Reported-by: NStanislaw Gruszka <sgruszka@redhat.com>
Signed-off-by: NShyam Iyer <shyam_iyer@dell.com>
Signed-off-by: NJon Mason <mason@myri.com>
Acked-by: NJesse Barnes <jbarnes@virtuousgeek.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>