* pci/bjorn-p2p-bridge-windows:
sparc/PCI: replace pci_cfg_fake_ranges() with pci_read_bridge_bases()
PCI: support sizing P2P bridge I/O windows with 1K granularity
PCI: reimplement P2P bridge 1K I/O windows (Intel P64H2)
PCI: allow P2P bridge windows starting at PCI bus address zero
Conflicts:
drivers/pci/probe.c
include/linux/pci.h
* pci/bjorn-disable-decode:
PCI: disable MEM decoding while updating 64-bit MEM BARs
PCI: leave MEM and IO decoding disabled during 64-bit BAR sizing, too
9d265124d0 and 15a260d53f added quirks for P2P bridges that support
I/O windows that start/end at 1K boundaries, not just the 4K boundaries
defined by the PCI spec. For details, see the IOBL_ADR register and the
EN1K bit in the CNF register in the Intel 82870P2 (P64H2).
These quirks complicate the code that reads P2P bridge windows
(pci_read_bridge_io() and pci_cfg_fake_ranges()) because the bridge
I/O resource is updated in the HEADER quirk, in pci_read_bridge_io(),
in pci_setup_bridge(), and again in the FINAL quirk. This is confusing
and makes it impossible to reassign the bridge windows after FINAL
quirks are run.
This patch adds support for 1K windows in the generic paths, so the
HEADER quirk only has to enable this support. The FINAL quirk, which
used to undo damage done by pci_setup_bridge(), is no longer needed.
This removes "if (!res->start) res->start = ..." from pci_read_bridge_io();
that was part of 9d265124d0 to avoid overwriting the resource filled in
by the quirk. Since pci_read_bridge_io() itself now knows about
granularity, the quirk no longer updates the resource and this test is no
longer needed.
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
After 253d2e5498, we disable MEM and IO decoding for most devices while we
size 32-bit BARs. However, we restore the original COMMAND register before
we size the upper 32 bits of 64-bit BARs, so we can still cause a conflict.
This patch waits to restore the original COMMAND register until we're
completely finished sizing the BAR.
Reference: https://lkml.org/lkml/2007/8/25/154
Acked-by: Jacob Pan <jacob.jun.pan@linux.intel.com>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
cd81e1ea1a added checks that prevent us from using P2P bridge windows
that start at PCI bus address zero. The reason was to "prevent us from
overwriting resources that are unassigned."
But generic code should allow address zero in both BARs and bridge
windows, so I think that commit was a mistake.
Windows at bus address zero are legal and likely to exist on machines with
an offset between bus addresses and CPU addresses. For example, in the
following hypothetical scenario, the bridge at 00:01.0 has a window at bus
address zero and the device at 01:00.0 has a BAR at bus address zero, and
I think both are perfectly valid:
PCI host bridge to bus 0000:00
pci_bus 0000:00: root bus resource [mem 0x100000000-0x1ffffffff] (bus address [0x00000000-0xffffffff])
pci 0000:00:01.0: PCI bridge to [bus 01]
pci 0000:00:01.0: bridge window [mem 0x100000000-0x100ffffff]
pci 0000:01:00.0: reg 10: [mem 0x100000000-0x100ffffff]
Acked-by: Yinghai Lu <yinghai@kernel.org>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
On P2P bridges with 32-bit I/O decoding, we incorrectly sign-extended
windows starting at 0x80000000 or above. In "base |= (io_base_hi << 16)",
"io_base_hi" is promoted to a signed int before being extended to an
unsigned long.
This would cause a window starting at I/O address 0x80000000 to be
treated as though it started at 0xffffffff80008000 instead, which
should cause "no compatible bridge window" errors when we enumerate
devices using that I/O space.
The mmio and mmio_pref casts are not strictly necessary, but without
them, correctness depends on the types of the PCI_MEMORY_RANGE_MASK and
PCI_PREF_RANGE_MASK constants, which are not obvious from reading the
local code.
Found by Coverity (CID 138747 and CID 138748).
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Now we can insert busn_res now, after all root bus's get inserted.
Signed-off-by: Yinghai Lu <yinghai@kernel.org>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
We need to put into the resources list for legacy system.
Signed-off-by: Yinghai Lu <yinghai@kernel.org>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Some callers do not supply the bus number aperture, usually because they do
not know the end. In this case, we assume the aperture extends from the
root bus number to bus 255, scan the bus, and shrink the bus number
resource so it ends at the largest bus number we found.
This is obviously not correct because the actual end of the aperture may
well be larger than the largest bus number we found. But I guess it's all
we have for now.
Also print out one info about that, so we could find out which path
does not have busn_res in resources list.
[bhelgaas: changelog, _safe iterator unnecessary, use %pR format for bus]
Signed-off-by: Yinghai Lu <yinghai@kernel.org>
Will use them insert/update busn res in pci_bus struct.
[bhelgaas: print conflicting entry if insertion fails]
Signed-off-by: Yinghai Lu <yinghai@kernel.org>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
This adds get_pci_domain_busn_res(), which returns the root of the
bus number resource tree for a domain, creating it if necessary.
We will later populate the tree with the bus numbers used by host
bridges and P2P bridges in the domain.
[bhelgaas: changelog]
Signed-off-by: Yinghai Lu <yinghai@kernel.org>
Signed-off-by: Bjorn 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: Yinghai Lu <yinghai@kernel.org>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
A PCIe downstream port is a P2P bridge. Its secondary interface is
a link that should lead only to device 0 (unless ARI is enabled)[1], so
we don't probe for non-zero device numbers.
Some Stratus ftServer systems have a PCIe downstream port (02:00.0) that
leads to both an upstream port (03:00.0) and a downstream port (03:01.0),
and 03:01.0 has important devices below it:
[0000:02]-+-00.0-[03-3c]--+-00.0-[04-09]--...
\-01.0-[0a-0d]--+-[USB]
+-[NIC]
+-...
Previously, we didn't enumerate device 03:01.0, so USB and the network
didn't work. This patch adds a DMI quirk to scan all device numbers,
not just 0, below a downstream port.
Based on a patch by Prarit Bhargava.
[1] PCIe spec r3.0, sec 7.3.1
CC: Myron Stowe <mstowe@redhat.com>
CC: Don Dutile <ddutile@redhat.com>
CC: James Paradis <james.paradis@stratus.com>
CC: Matthew Wilcox <matthew.r.wilcox@intel.com>
CC: Jesse Barnes <jbarnes@virtuousgeek.org>
CC: Prarit Bhargava <prarit@redhat.com>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
We need a hook to release host bridge resources allocated when creating
root bus.
Signed-off-by: Yinghai Lu <yinghai@kernel.org>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Use that device for pci_root_bus bridge pointer.
Use pci_release_bus_bridge_dev() to release allocated pci_host_bridge in
remove path.
Use root bus bridge pointer to get host bridge pointer instead of searching
host bridge list. That leaves the host bridge list unused, so remove it.
Signed-off-by: Yinghai Lu <yinghai@kernel.org>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Move host bridge-related code from probe.c to a new host-bridge.c.
Signed-off-by: Yinghai Lu <yinghai@kernel.org>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
In 5bfa14ed9f, I forgot to initialize res2.flags before calling
pcibios_bus_to_resource(), which depends on the resource type to locate the
correct aperture. This bug won't hurt x86, which currently never has an
offset between bus and CPU addresses, but will affect other architectures.
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
This isn't really a quirk; calling it directly from pci_add_device makes
more sense.
Signed-off-by: Yinghai Lu <yinghai@kernel.org>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
Everybody uses the generic pcibios_resource_to_bus() supplied by the core
now, so remove the ARCH_HAS_GENERIC_PCI_OFFSETS used during conversion.
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
This replaces the generic versions of pcibios_resource_to_bus() and
pcibios_bus_to_resource() in asm-generic/pci.h with versions that use
pci_resource_to_bus() and pci_bus_to_resource().
The replacements are equivalent except that they can apply host
bridge window offsets when the arch has supplied them by using
pci_add_resource_offset().
Each arch can convert to using pci_add_resource_offset() individually by
removing its device resource fixups from pcibios_fixup_bus() and supplying
ARCH_HAS_GENERIC_PCI_OFFSETS. ARCH_HAS_GENERIC_PCI_OFFSETS can be removed
after all have converted.
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Some PCI host bridges translate CPU addresses to PCI bus addresses.
Previously, we initialized pci_dev resources with PCI bus addresses,
then converted them to CPU addresses later in arch-specific code
(pcibios_fixup_resources()), which leaves a window of time where the
pci_dev resources are incorrect.
This patch adds support in the core for this address translation.
When the arch creates the root bus, it can supply the host bridge
address translation information, and the core can use it to set the
pci_dev resources correctly from the beginning.
This gives us a way to fix the problem that quirks that run between device
discovery and pcibios_fixup_resources() fail because they use pci_dev
resources that haven't been converted. The reference below is to one
such problem that affected ARM and ia64.
Note that this patch has no effect until an arch starts using
pci_add_resource_offset() with a non-zero offset: before that, all
all host bridge windows have a zero offset and pci_bus_to_resource()
copies the pci_bus_region directly to the struct resource.
Reference: https://lkml.org/lkml/2009/10/12/405
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Some PCI host bridges apply an address offset, so bus addresses on PCI are
different from CPU addresses. This patch adds a way for architectures to
tell the PCI core about this offset. For example:
LIST_HEAD(resources);
pci_add_resource_offset(&resources, host->io_space, host->io_offset);
pci_add_resource_offset(&resources, host->mem_space, host->mem_offset);
pci_scan_root_bus(parent, bus, ops, sysdata, &resources);
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
This adds a list of all PCI host bridges we find and a way to look up
the host bridge from a pci_dev.
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
When pci_create_root_bus() adds the new struct pci_bus to the global
pci_root_buses list, the bus becomes visible to other parts of the
kernel, so it should be fully initialized.
This patch delays adding the bus to the pci_root_buses list until after
all the struct pci_bus initialization is finished.
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Found debug print of class is shifted.
| pci 0000:f8:15.2: [8086:2b56] type 0 class 0x000600
Code is trying to print class with 6 digits, but use shifted class with
4 digits valid value as variable.
Change to original dev->class directly.
Also remove not needed calculating of local variable class, because it
will be updated after pci_fixup_device(pci_fixup_early...)
Also unify type print out when class and header is not matched.
Signed-off-by: Yinghai Lu <yinghai@kernel.org>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
We can reuse it for pciehp probing.
-v2: according to Kenji, fix crs timeout checking, and export the function
for later use when pciehp is compiled as a module.
Suggested-by: Matthew Wilcox <matthew@wil.cx>
Signed-off-by: Yinghai Lu <yinghai@kernel.org>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
This allows us to allocate resources to hotplug bridges during
remove/rescan.
We need to move the function to setup-bus.c so it can use
__pci_bus_size_bridges and __pci_bus_assign_resources directly to take
the add_list resource tracking list.
Signed-off-by: Yinghai Lu <yinghai@kernel.org>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
Current rescan will not touch bridge MMIO and IO.
Try to reuse pci_assign_unassigned_bridge_resources(bridge) to update bridge
resources, if child devices need more resources.
Only do that for bridges whose children are all removed already; i.e. don't
release resources that could already be in use by drivers on child devices.
Signed-off-by: Yinghai Lu <yinghai@kernel.org>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
Fixes PCI device detection on IBM xSeries IBM 3850 M2 / x3950 M2
when using ACPI resources (_CRS).
This is default, a manual workaround (without this patch)
would be pci=nocrs boot param.
V2: Add dev_warn if the workaround is hit. This should reveal
how common such setups are (via google) and point to possible
problems if things are still not working as expected.
-> Suggested by Jan Beulich.
Cc: stable@vger.kernel.org
Tested-by: garyhade@us.ibm.com
Signed-off-by: Yinghai Lu <yinghai.lu@oracle.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
All users of pci_create_bus() have been converted to pci_create_root_bus(),
so remove it.
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
Users of pci_scan_bus_parented() should be converted to use either
pci_scan_root_bus() (preferred, but also calls pci_bus_add_devices)
or
pci_create_root_bus()
pci_scan_child_bus()
Since pci_scan_bus_parented(), I'm marking it deprecated now and will
actually remove it later.
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
This converts pci_scan_bus_parented() to use pci_create_root_bus()
instead of pci_create_bus(). The new bus still has the default (incorrect)
resources, so this patch doesn't help fix that problem, but it does remove
one more use of pci_create_bus().
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
I plan to deprecate pci_scan_bus_parented(), so use pci_create_root_bus()
directly instead. pci_scan_bus() itself will be removed as soon as all
callers are gone, so this is just an interim step.
v2: export pci_scan_bus
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
"Early" and "header" quirks often use incorrect bus resources because they
see the default resources assigned by pci_create_bus(), before the
architecture fixes them up (typically in pcibios_fixup_bus()). Regions
reserved by these quirks end up with the wrong parents.
Here's the standard path for scanning a PCI root bus:
pci_scan_bus or pci_scan_bus_parented
pci_create_bus <-- A create with default resources
pci_scan_child_bus
pci_scan_slot
pci_scan_single_device
pci_scan_device
pci_setup_device
pci_fixup_device(early) <-- B
pci_device_add
pci_fixup_device(header) <-- C
pcibios_fixup_bus <-- D fill in correct resources
Early and header quirks at B and C use the default (incorrect) root bus
resources rather than those filled in at D.
This patch adds a new pci_scan_root_bus() function that sets the bus
resources correctly from a supplied list of resources.
I intend to remove pci_scan_bus() and pci_scan_bus_parented() after
fixing all callers.
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
pci_create_bus() assigns ioport_resource and iomem_resource as the default
bus resources, i.e., the entire address space. Architectures fix these
later, typically in pcibios_fixup_bus() or after pci_scan_bus_parented()
returns, but code that runs in the interim sees incorrect resource
information.
This patch adds a new pci_create_root_bus() that sets the bus resources
correctly from a supplied list of resources.
I intend to remove pci_create_bus() after changing all callers.
Based on original patch by Deng-Cheng Zhu.
Reference: http://www.spinics.net/lists/mips/msg41654.html
Reference: https://lkml.org/lkml/2011/8/26/88
Signed-off-by: Deng-Cheng Zhu <dczhu@mips.com>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
Show the bus number and resources for every root bus we create. This
will become more interesting when we supply the correct resources
instead of using the defaults (ioport_resource and iomem_resource).
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
Clean-up MPS debug output to make it a single line and aligned, thus
making it more readable for a large number of buses and devices in a
single system.
Suggested by Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Jon Mason <mason@myri.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
Rework the "performance" MPS option to configure the device MPS with the
smaller of the device MPSS or the bridge MPS (which is assumed to be
properly configured at this point to the largest allowable MPS based on
its parent bus).
Also, rework the MRRS setting to report an inability to set the MRRS to
a valid setting.
Signed-off-by: Jon Mason <mason@myri.com>
Acked-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
Add the ability to disable PCI-E MPS turning and using the BIOS
configured MPS defaults. Due to the number of issues recently
discovered on some x86 chipsets, make this the default behavior.
Also, add the option for peer to peer DMA MPS configuration. Peer to
peer DMA is outside the scope of this patch, but MPS configuration could
prevent it from working by having the MPS on one root port different
than the MPS on another. To work around this, simply make the system
wide MPS the smallest possible value (128B).
Signed-off-by: Jon Mason <mason@myri.com>
Acked-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
In pcie_find_smpss(), we have the following statement:
if (dev->is_hotplug_bridge && (!list_is_singular(&dev->bus->devices) ||
dev->bus->self->pcie_type != PCI_EXP_TYPE_ROOT_PORT))
The problem is that at least on my machine, this gets called for the
root complex (virtual P2P bridge), and dev->bus->self is NULL since
the parent bus for this is not itself anchor to a PCI device.
This adds the necessary NULL check.
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Acked-by: Jon Mason <mason@myri.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Modifying the Maximum Read Request Size to 0 (value of 128Bytes) has
massive negative ramifications on some devices. Without knowing which
devices have this issue, do not modify from the default value when
walking the PCI-E bus in pcie_bus_safe mode. Also, make pcie_bus_safe
the default procedure.
Tested-by: Sven Schnelle <svens@stackframe.org>
Tested-by: Simon Kirby <sim@hostway.ca>
Tested-by: Stephen M. Cameron <scameron@beardog.cce.hp.com>
Reported-and-tested-by: Eric Dumazet <eric.dumazet@gmail.com>
Reported-and-tested-by: Niels Ole Salscheider <niels_ole@salscheider-online.de>
References: https://bugzilla.kernel.org/show_bug.cgi?id=42162
Signed-off-by: Jon Mason <mason@myri.com>
Acked-by: Jesse Barnes <jbarnes@virtuousgeek.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Commit b03e7495a8 ("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: Stanislaw Gruszka <sgruszka@redhat.com>
Signed-off-by: Shyam Iyer <shyam_iyer@dell.com>
Signed-off-by: Jon Mason <mason@myri.com>
Acked-by: Jesse Barnes <jbarnes@virtuousgeek.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
pcie_bus_configure_settings needs to be exported if the PCI hotplug
driver is being compiled as a module.
Reported-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Jon Mason <mason@myri.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
On a given PCI-E fabric, each device, bridge, and root port can have a
different PCI-E maximum payload size. There is a sizable performance
boost for having the largest possible maximum payload size on each PCI-E
device. However, if improperly configured, fatal bus errors can occur.
Thus, it is important to ensure that PCI-E payloads sends by a device
are never larger than the MPS setting of all devices on the way to the
destination.
This can be achieved two ways:
- A conservative approach is to use the smallest common denominator of
the entire tree below a root complex for every device on that fabric.
This means for example that having a 128 bytes MPS USB controller on one
leg of a switch will dramatically reduce performances of a video card or
10GE adapter on another leg of that same switch.
It also means that any hierarchy supporting hotplug slots (including
expresscard or thunderbolt I suppose, dbl check that) will have to be
entirely clamped to 128 bytes since we cannot predict what will be
plugged into those slots, and we cannot change the MPS on a "live"
system.
- A more optimal way is possible, if it falls within a couple of
constraints:
* The top-level host bridge will never generate packets larger than the
smallest TLP (or if it can be controlled independently from its MPS at
least)
* The device will never generate packets larger than MPS (which can be
configured via MRRS)
* No support of direct PCI-E <-> PCI-E transfers between devices without
some additional code to specifically deal with that case
Then we can use an approach that basically ignores downstream requests
and focuses exclusively on upstream requests. In that case, all we need
to care about is that a device MPS is no larger than its parent MPS,
which allows us to keep all switches/bridges to the max MPS supported by
their parent and eventually the PHB.
In this case, your USB controller would no longer "starve" your 10GE
Ethernet and your hotplug slots won't affect your global MPS.
Additionally, the hotplugged devices themselves can be configured to a
larger MPS up to the value configured in the hotplug bridge.
To choose between the two available options, two PCI kernel boot args
have been added to the PCI calls. "pcie_bus_safe" will provide the
former behavior, while "pcie_bus_perf" will perform the latter behavior.
By default, the latter behavior is used.
NOTE: due to the location of the enablement, each arch will need to add
calls to this function. This patch only enables x86.
This patch includes a number of changes recommended by Benjamin
Herrenschmidt.
Tested-by: Jordan_Hargrave@dell.com
Signed-off-by: Jon Mason <mason@myri.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>