Add ids module parameter which allows specifying initial IDs for the
pci-stub driver.  When built into the kernel, pci-stub is linked
before any real pci drivers and by setting up IDs from initialization
it can prevent built-in drivers from attaching to specific devices.

While at it, make pci_stub_probe() print out about devices it grabbed
to weed out "but my controller isn't being probed" bug reports.
Signed-off-by: NTejun Heo <tj@kernel.org>
Signed-off-by: NJesse Barnes <jbarnes@virtuousgeek.org>

Separate out pci_add_dynid() from store_new_id() and export it so that
in-kernel code can add PCI IDs dynamically.  As the function will be
available regardless of HOTPLUG, put it and pull pci_free_dynids()
outside of CONFIG_HOTPLUG.

This will be used by pci-stub to initialize initial IDs via module
param.

While at it, remove bogus get_driver() failure check.
Signed-off-by: NTejun Heo <tj@kernel.org>
Acked-by: NGreg Kroah-Hartman <gregkh@suse.de>
Reviewed-by: NGrant Grundler <grundler@parisc-linux.org>
Signed-off-by: NJesse Barnes <jbarnes@virtuousgeek.org>

Add support for PCI-E 5.0 GT/s in max_bus_speed and cur_bus_speed.
Reviewed-by: NMatthew Wilcox <matthew@wil.cx>
Signed-off-by: NKenji Kaneshige <kaneshige.kenji@jp.fujitsu.com>
Signed-off-by: NJesse Barnes <jbarnes@virtuousgeek.org>

Fix typo in PCI-E link speed.
Reviewed-by: NMatthew Wilcox <matthew@wil.cx>
Signed-off-by: NKenji Kaneshige <kaneshige.kenji@jp.fujitsu.com>
Signed-off-by: NJesse Barnes <jbarnes@virtuousgeek.org>

The L0s state can be managed separately for each direction (upstream
direction and downstream direction) of the link. But in the current
implementation, those are mixed up. With this patch, L0s for each
direction are managed separately.

To maintain three states (upstream direction L0s, downstream L0s and
L1), 'aspm_support', 'aspm_enabled', 'aspm_capable', 'aspm_disable'
and 'aspm_default' fields in struct pcie_link_state are changed to
3-bit from 2-bit. The 'latency' field is separated to two 'latency_up'
and 'latency_dw' fields to maintain exit latencies for each direction
of the link. For L0, 'latency_up.l0' and 'latency_dw.l0' are used to
configure upstream direction L0s and downstream direction L0s
respectively. For L1, larger value of 'latency_up.l1' and
'latency_dw.l1' is considered as L1 exit latency.
Acked-by: NShaohua Li <shaohua.li@intel.com>
Signed-off-by: NKenji Kaneshige <kaneshige.kenji@jp.fujitsu.com>
Signed-off-by: NJesse Barnes <jbarnes@virtuousgeek.org>

In the current implementation, ASPM L0s/L1 is disabled for all links
in the hierarchy if one of the link doesn't meet latency requirement.
But we can partially enable ASPM L0s/L1 on sub-tree in the hierarchy.
This patch allows partial L0s/L1 enablement in the hierarchy. And it
also reduce the calculation cost of ASPM configuration very much.

In the previous implementation, all links were enabled with the same
state. With this patch, enabled state for each link is determined
simply as follows (the 'requested' is from policy_to_aspm_state()).

    enabled = requested & (link->aspm_capable & link->aspm_disable)
Acked-by: NShaohua Li <shaohua.li@intel.com>
Signed-off-by: NKenji Kaneshige <kaneshige.kenji@jp.fujitsu.com>
Signed-off-by: NJesse Barnes <jbarnes@virtuousgeek.org>

Introduce 'aspm_capable' field to maintain the capable ASPM setting of
the link. By the 'aspm_capable', we don't need to recheck latency
every time ASPM policy is changed.

Each bit in 'aspm_capable' is associated to ASPM state (L0S/L1). The
bit is set if the associated ASPM state is supported by the link and
it satisfies the latency requirement (i.e. exit latency < endpoint
acceptable latency). The 'aspm_capable' is updated when

  - an endpoint device is added (boot time or hot-plug time)
  - an endpoint device is removed (hot-unplug time)
  - PCI power state is changed.
Acked-by: NShaohua Li <shaohua.li@intel.com>
Signed-off-by: NKenji Kaneshige <kaneshige.kenji@jp.fujitsu.com>
Signed-off-by: NJesse Barnes <jbarnes@virtuousgeek.org>

Introduce 'aspm_disable' flag to manage disabled ASPM state more
robust way.
Acked-by: NShaohua Li <shaohua.li@intel.com>
Signed-off-by: NKenji Kaneshige <kaneshige.kenji@jp.fujitsu.com>
Signed-off-by: NJesse Barnes <jbarnes@virtuousgeek.org>

Fix possible NULL dereference in pcie_aspm_exit_link_state(). This
patch also cleanup some code.
Acked-by: NShaohua Li <shaohua.li@intel.com>
Signed-off-by: NKenji Kaneshige <kaneshige.kenji@jp.fujitsu.com>
Signed-off-by: NJesse Barnes <jbarnes@virtuousgeek.org>

Remove the following check in __pcie_aspm_config_link() because it
nerver be true.
Acked-by: NShaohua Li <shaohua.li@intel.com>
Signed-off-by: NKenji Kaneshige <kaneshige.kenji@jp.fujitsu.com>
Signed-off-by: NJesse Barnes <jbarnes@virtuousgeek.org>

We must not clear bits in 'aspm_enabled' using 'aspm_support', or
'aspm_enabled' and 'aspm_default' might be different from the actual
state. In addtion, 'aspm_default' should be intialized even if
'aspm_support' is 0.
Acked-by: NShaohua Li <shaohua.li@intel.com>
Signed-off-by: NKenji Kaneshige <kaneshige.kenji@jp.fujitsu.com>
Signed-off-by: NJesse Barnes <jbarnes@virtuousgeek.org>

This patch adds the most recent additions to the list of 82576 device IDs
to the list of devices needing the SR-IOV quirk.
Signed-off-by: NAlexander Duyck <alexander.h.duyck@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: NJesse Barnes <jbarnes@virtuousgeek.org>

Background:
Graphic devices are accessed through ranges in I/O or memory space. While most
modern devices allow relocation of such ranges, some "Legacy" VGA devices
implemented on PCI will typically have the same "hard-decoded" addresses as
they did on ISA. For more details see "PCI Bus Binding to IEEE Std 1275-1994
Standard for Boot (Initialization Configuration) Firmware Revision 2.1"
Section 7, Legacy Devices.

The Resource Access Control (RAC) module inside the X server currently does
the task of arbitration when more than one legacy device co-exists on the same
machine. But the problem happens when these devices are trying to be accessed
by different userspace clients (e.g. two server in parallel). Their address
assignments conflict. Therefore an arbitration scheme _outside_ of the X
server is needed to control the sharing of these resources. This document
introduces the operation of the VGA arbiter implemented for Linux kernel.
Signed-off-by: NBenjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: NTiago Vignatti <tiago.vignatti@nokia.com>
Signed-off-by: NDave Airlie <airlied@redhat.com>
Signed-off-by: NJesse Barnes <jbarnes@virtuousgeek.org>

Cleanups (nearly based on checkpatch).

Before: total: 11 errors, 2 warnings, 0 checks, 842 lines checked
After:  total:  0 errors, 0 warnings, 0 checks, 842 lines checked

v2: fix it's/its mistakes in comment
Reviewed-by: NMatthew Wilcox <willy@linux.intel.com>
Signed-off-by: NHidetoshi Seto <seto.hidetoshi@jp.fujitsu.com>
Signed-off-by: NJesse Barnes <jbarnes@virtuousgeek.org>

Cleanup based on the prototype from Matthew Milcox.
Reviewed-by: NMatthew Wilcox <willy@linux.intel.com>
Signed-off-by: NHidetoshi Seto <seto.hidetoshi@jp.fujitsu.com>
Signed-off-by: NJesse Barnes <jbarnes@virtuousgeek.org>

Cleanup based on the prototype from Matthew Milcox.
Reviewed-by: NMatthew Wilcox <willy@linux.intel.com>
Signed-off-by: NHidetoshi Seto <seto.hidetoshi@jp.fujitsu.com>
Signed-off-by: NJesse Barnes <jbarnes@virtuousgeek.org>

Cleanup based on the prototype from Matthew Milcox.
Reviewed-by: NMatthew Wilcox <willy@linux.intel.com>
Signed-off-by: NHidetoshi Seto <seto.hidetoshi@jp.fujitsu.com>
Signed-off-by: NJesse Barnes <jbarnes@virtuousgeek.org>

Move it from the middle of the function to the end.
Reviewed-by: NMatthew Wilcox <willy@linux.intel.com>
Signed-off-by: NHidetoshi Seto <seto.hidetoshi@jp.fujitsu.com>
Signed-off-by: NJesse Barnes <jbarnes@virtuousgeek.org>

Unify msi_free_irqs() and msix_free_all_irqs(), and rename it to a
common void function free_msi_irqs().

And relocate the common function to where the prototype is located now.
Reviewed-by: NMatthew Wilcox <willy@linux.intel.com>
Signed-off-by: NHidetoshi Seto <seto.hidetoshi@jp.fujitsu.com>
Signed-off-by: NJesse Barnes <jbarnes@virtuousgeek.org>

use list_first_entry() instead of list_entry().
Reviewed-by: NMatthew Wilcox <willy@linux.intel.com>
Signed-off-by: NHidetoshi Seto <seto.hidetoshi@jp.fujitsu.com>
Signed-off-by: NJesse Barnes <jbarnes@virtuousgeek.org>

The msi_list never have MSI-X's msi_desc while MSI is enabled,
and also it never have MSI's msi_desc while MSI-X is enabled.

This patch remove check for MSI-X entry from the pci_disable_msi(),
referring that pci_disable_msix() does not have any check for MSI
entry.
Reviewed-by: NMatthew Wilcox <willy@linux.intel.com>
Signed-off-by: NHidetoshi Seto <seto.hidetoshi@jp.fujitsu.com>
Signed-off-by: NJesse Barnes <jbarnes@virtuousgeek.org>

We already print it out for pci bridges, so also print it out for pci devices.
Signed-off-by: NYinghai Lu <yinghai@kernel.org>
Signed-off-by: NJesse Barnes <jbarnes@virtuousgeek.org>

http://bugzilla.kernel.org/show_bug.cgi?id=12542 reports that with the
quirk not applied on resume, msi stops working after resuming and mcp78s
ahci fails due to IRQ mis-delivery.  Apply it on resume too.
Signed-off-by: NTejun Heo <tj@kernel.org>
Cc: Peer Chen <pchen@nvidia.com>
Cc: Tj <linux@tjworld.net>
Reported-by: NNicolas Derive <kalon33@ubuntu.com>
Cc: Greg KH <greg@kroah.com>
Cc: <stable@kernel.org>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NJesse Barnes <jbarnes@virtuousgeek.org>

Shut off the long standing

linux/drivers/pci/search.c:144: warning: 'pci_find_device' is deprecated (declared at linux/drivers/pci/search.c:136)
linux/drivers/pci/search.c:144: warning: 'pci_find_device' is deprecated (declared at linux/drivers/pci/search.c:136)

warnings that appear on every build when CONFIG_PCI_LEGACY is enabled.

gcc warns for the use in EXPORT_SYMBOL

I moved these to a separate file and disabled the warning in the Makefile for that file.
Signed-off-by: NAndi Kleen <ak@linux.intel.com>
Signed-off-by: NJesse Barnes <jbarnes@virtuousgeek.org>

One more form factor for Compaq Evo D510, which needs the same quirk
as the other form factors. Apparently there's no hardware monitoring
chip on that one, but SPD EEPROMs, so it's still worth unhiding the
SMBus.
Signed-off-by: NJean Delvare <khali@linux-fr.org>
Tested-by: NNuzhna Pomoshch <nuzhna_pomoshch@yahoo.com>
Signed-off-by: NJesse Barnes <jbarnes@virtuousgeek.org>

Some devices allow an individual function to be reset without affecting
other functions in the same device: that's what pci_reset_function does.
For devices that have this support, expose reset attribite in sysfs.

This is useful e.g. for virtualization, where a qemu userspace
process wants to reset the device when the guest is reset,
to emulate machine reboot as closely as possible.
Acked-by: NGreg Kroah-Hartman <gregkh@suse.de>
Signed-off-by: NMichael S. Tsirkin <mst@redhat.com>
Signed-off-by: NJesse Barnes <jbarnes@virtuousgeek.org>

We cannot simply call acpi_get_pci_dev() on any random ACPI handle
and hope that it works, because a PCI root bridge may not have
an associated struct pci_dev.

This is allowed per the PCI specification, and is referred to as a
non-materialized bridge.

So, depending on the type of PCI bridge that the handle points to,
use the appropriate interface to return the struct pci_bus correctly.
Reviewed-by: NBjorn Helgaas <bjorn.helgaas@hp.com>
Signed-off-by: NAlex Chiang <achiang@hp.com>
Signed-off-by: NJesse Barnes <jbarnes@virtuousgeek.org>

yenta needs this for example.
Acked-by: NMatthew Wilcox <willy@linux.intel.com>
Reported-by: NStephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: NJesse Barnes <jbarnes@virtuousgeek.org>

This was #define'd as 0 on all platforms, so let's get rid of it.

This change makes pci_scan_slot() slightly easier to read.

Cc: Yoshinori Sato <ysato@users.sourceforge.jp>
Cc: Tony Luck <tony.luck@intel.com>
Cc: David Howells <dhowells@redhat.com>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: Jeff Dike <jdike@addtoit.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Ivan Kokshaysky <ink@jurassic.park.msu.ru>
Reviewed-by: NMatthew Wilcox <willy@linux.intel.com>
Acked-by: NRussell King <linux@arm.linux.org.uk>
Acked-by: NRalf Baechle <ralf@linux-mips.org>
Acked-by: NKyle McMartin <kyle@mcmartin.ca>
Acked-by: NBenjamin Herrenschmidt <benh@kernel.crashing.org>
Acked-by: NPaul Mundt <lethal@linux-sh.org>
Acked-by: NArnd Bergmann <arnd@arndb.de>
Signed-off-by: NAlex Chiang <achiang@hp.com>
Signed-off-by: NJesse Barnes <jbarnes@virtuousgeek.org>

An SR-IOV capable device includes an SR-IOV PCIe capability which
describes the Virtual Function (VF) BAR requirements.  A typical SR-IOV
device can support multiple VFs whose BARs must be in a contiguous region,
effectively an array of VF BARs.  The BAR reports the size requirement
for a single VF.  We calculate the full range needed by simply multiplying
the VF BAR size with the number of possible VFs and create a resource
spanning the full range.

This all seems sane enough except it artificially inflates the alignment
requirement for the VF BAR.  The VF BAR need only be aligned to the size
of a single BAR not the contiguous range of VF BARs.  This can cause us
to fail to allocate resources for the BAR despite the fact that we
actually have enough space.

This patch adds a thin PCI specific layer over the generic
resource_alignment() function which is aware of the special nature of
VF BARs and does sorting and allocation based on the smaller alignment
requirement.

I recognize that while resource_alignment is generic, it's basically a
PCI helper.  An alternative to this patch is to add PCI VF BAR specific
information to struct resource.  I opted for the extra layer rather than
adding such PCI specific information to struct resource.  This does
have the slight downside that we don't cache the BAR size and re-read
for each alignment query (happens a small handful of times during boot
for each VF BAR).
Signed-off-by: NChris Wright <chrisw@sous-sol.org>
Cc: Ivan Kokshaysky <ink@jurassic.park.msu.ru>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Matthew Wilcox <matthew@wil.cx>
Cc: Yu Zhao <yu.zhao@intel.com>
Cc: stable@kernel.org
Signed-off-by: NJesse Barnes <jbarnes@virtuousgeek.org>

Without the check, the config space may be filled with zeros. Though
the driver should try to avoid call restoring before saving, but the
pci layer also should check this.

Also removes the existing check in pci_restore_standard_config, since
it's superfluous with the new check in restore_state.
Acked-by: NRafael J. Wysocki <rjw@sisk.pl>
Signed-off-by: NAlek Du <alek.du@intel.com>
Signed-off-by: NJesse Barnes <jbarnes@virtuousgeek.org>

By the pci slot changes, callbacks of attributes under slot directory
(/sys/bus/pci/slots) had been changed to get the pointer to struct
pci_slot instead of struct hotplug_slot. So the path_show() that
assumes the parameter is a pointer to struct hotplug_slot seems
broken.
Tested-by: NMike Habeck <habeck@sgi.com>
Signed-off-by: NKenji Kaneshige <kaneshige.kenji@jp.fujitsu.com>
Signed-off-by: NJesse Barnes <jbarnes@virtuousgeek.org>

The commit bd3d99c1 ("PCI: Remove
untested Electromechanical Interlock (EMI) support in pciehp."), which
removes the definition of "struct hotplug_slot_attr", broke SGI
hotplug driver. By this commit, we get the following compile error.

drivers/pci/hotplug/sgi_hotplug.c:106: error: variable 'sn_slot_path_attr' has initializer but incomplete type
drivers/pci/hotplug/sgi_hotplug.c:106: error: unknown field 'attr' specified in initializer
drivers/pci/hotplug/sgi_hotplug.c:106: error: extra brace group at end of initializer
drivers/pci/hotplug/sgi_hotplug.c:106: error: (near initialization for 'sn_slot_path_attr')
drivers/pci/hotplug/sgi_hotplug.c:106: warning: excess elements in struct initializer
drivers/pci/hotplug/sgi_hotplug.c:106: warning: (near initialization for 'sn_slot_path_attr')
drivers/pci/hotplug/sgi_hotplug.c:106: error: unknown field 'show' specified in initializer
drivers/pci/hotplug/sgi_hotplug.c:106: warning: excess elements in struct initializer
drivers/pci/hotplug/sgi_hotplug.c:106: warning: (near initialization for 'sn_slot_path_attr')
drivers/pci/hotplug/sgi_hotplug.c: In function 'sn_hp_destroy':
drivers/pci/hotplug/sgi_hotplug.c:203: error: invalid use of undefined type 'struct hotplug_slot_attribute'
drivers/pci/hotplug/sgi_hotplug.c: In function 'sn_hotplug_slot_register':
drivers/pci/hotplug/sgi_hotplug.c:655: error: invalid use of undefined type 'struct hotplug_slot_attribute'

This patch fixes this regression by adding the definition of struct
hotplug_slot_attr into sgi_hotplug.c.
Tested-by: NMike Habeck <habeck@sgi.com>
Signed-off-by: NKenji Kaneshige <kaneshige.kenji@jp.fujitsu.com>
Signed-off-by: NJesse Barnes <jbarnes@virtuousgeek.org>

Two defects work together result in KVM device passthrough randomly can't
work:
1. iommu_snooping is not initialized to zero when vm_iommu_init() called.
So it is possible to get a random value.
2. One line added by commit 2c2e2c38("IOMMU Identity Mapping Support")
change the code path, let it bypass domain_update_iommu_cap(), as well as
missing the increment of domain iommu reference count.

The latter is also likely to cause a leak of domains on repeated VMM 
assignment and deassignment.
Signed-off-by: NSheng Yang <sheng@linux.intel.com>
Signed-off-by: NDavid Woodhouse <David.Woodhouse@intel.com>

The physical address passed to domain_pfn_mapping() should be rounded 
down to the start of the MM page, not the VT-d page.

This issue causes kernel panic on PAGE_SIZE>VTD_PAGE_SIZE platforms e.g. ia64
platforms.
Signed-off-by: NFenghua Yu <fenghua.yu@intel.com>
Signed-off-by: NDavid Woodhouse <David.Woodhouse@intel.com>

In domain_sg_mapping(), use aligned_nrpages() instead of hand-coded
rounding code for calculating the size of each sg elem. This means that
on IA64 we correctly round up to the MM page size, not just to the VT-d
page size.

Also remove the incorrect mm_to_dma_pfn() when intel_map_sg() calls
domain_sg_mapping() -- the 'size' variable is in VT-d pages already.
Signed-off-by: NFenghua Yu <fenghua.yu@intel.com>
Signed-off-by: NDavid Woodhouse <David.Woodhouse@intel.com>

This function has traditionally used "insert_resource()", because before
commit cebd78a8 ("Fix pci_claim_resource") it used to just insert the
resource into whatever root resource tree that was indicated by
"pcibios_select_root()".

So there Matthew fixed it to actually look up the proper parent
resource, which means that now it's actively wrong to then traverse the
resource tree any more: we already know exactly where the new resource
should go.

And when we then did commit a76117df ("x86: Use pci_claim_resource"),
which changed the x86 PCI code from the open-coded

	pr = pci_find_parent_resource(dev, r);
	if (!pr || request_resource(pr, r) < 0) {

to using

	if (pci_claim_resource(dev, idx) < 0) {

that "insert_resource()" now suddenly became a problem, and causes a
regression covered by

	http://bugzilla.kernel.org/show_bug.cgi?id=13891

which this fixes.
Reported-and-tested-by: NRafael J. Wysocki <rjw@sisk.pl>
Cc: Matthew Wilcox <willy@linux.intel.com>
Cc: Andrew Patterson <andrew.patterson@hp.com>
Cc: Linux PCI <linux-pci@vger.kernel.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

* Remove smp_lock.h from files which don't need it (including some headers!)
* Add smp_lock.h to files which do need it
* Make smp_lock.h include conditional in hardirq.h
  It's needed only for one kernel_locked() usage which is under CONFIG_PREEMPT

  This will make hardirq.h inclusion cheaper for every PREEMPT=n config
  (which includes allmodconfig/allyesconfig, BTW)
Signed-off-by: NAlexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

After some API change, intel_iommu_unmap_range() introduced a assumption that
parameter size != 0, otherwise the dma_pte_clean_range() would have a
overflowed argument. But the user like KVM don't have this assumption before,
then some BUG() triggered.

Fix it by ignoring size = 0.
Signed-off-by: NSheng Yang <sheng@linux.intel.com>
Signed-off-by: NDavid Woodhouse <David.Woodhouse@intel.com>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Our current strategy for pass-through mode is to put all devices into
the 1:1 domain at startup (which is before we know what their dma_mask
will be), and only _later_ take them out of that domain, if it turns out
that they really can't address all of memory.

However, when there are a bunch of PCI devices behind a bridge, they all
end up with the same source-id on their DMA transactions, and hence in
the same IOMMU domain. This means that we _can't_ easily move them from
the 1:1 domain into their own domain at runtime, because there might be DMA
in-flight from their siblings.

So we have to adjust our pass-through strategy: For PCI devices not on
the root bus, and for the bridges which will take responsibility for
their transactions, we have to start up _out_ of the 1:1 domain, just in
case.

This fixes the BUG() we see when we have 32-bit-capable devices behind a
PCI-PCI bridge, and use the software identity mapping.

It does mean that we might end up using 'normal' mapping mode for some
devices which could actually live with the faster 1:1 mapping -- but
this is only for PCI devices behind bridges, which presumably aren't the
devices for which people are most concerned about performance.
Signed-off-by: NDavid Woodhouse <David.Woodhouse@intel.com>