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>
pcibios_scan_root() and pci_scan_bus_on_node() were almost identical,
so this patch merges them.
Signed-off-by: Yinghai Lu <yinghai@kernel.org>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
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: Bjorn Helgaas <bhelgaas@google.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
This doesn't change any functionality, but it makes a subsequent patch
slightly simpler.
pci_scan_bus(NULL, ...) and pci_scan_bus_parented() are identical except
that pci_scan_bus() also calls pci_bus_add_devices():
pci_scan_bus_parented
pci_create_bus
pci_scan_child_bus
pci_scan_bus
pci_create_bus
pci_scan_child_bus
pci_bus_add_devices
All callers of pcibios_scan_root() call pci_bus_add_devices() explicitly,
and we don't pass a parent device, so we might as well use pci_scan_bus().
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
As with any other such change, the goal is to prevent inadvertent
writes to these structures (assuming DEBUG_RODATA is enabled), and to
separate data (possibly frequently) written to from such never getting
modified.
Reviewed-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Jan Beulich <jbeulich@novell.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
Commit 6e8af08dfa enables pci=bfsort on
future Dell systems. But the identification string 'Dell System' matches
on already existing whitelist, which do not have SMBIOS type 0xB1,
causing pci=bfsort not being set on existing whitelist.
This patch fixes the regression by moving the type 0xB1 check beyond the
existing whitelist so that existing whitelist is walked before.
Signed-off-by: Shyam Iyer <shyam_iyer@dell.com>
Signed-off-by: Narendra K <narendra_k@dell.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
This patch enables pci=bfsort by default on future Dell systems.
It reads SMBIOS type 0xB1 vendor specific record and sets pci=bfsort
accordingly.
Offset Name Length Value Description
04 Flags0 Word Varies Bits 9-10
- 10:9 = 00 Unknown
- 10:9 = 01 Breadth First
- 10:9 = 10 Depth First
- 10:9 = 11 Reserved
1. Any time pci=bfsort has to be enabled on a system, we need to add the
model number of the system to the white list. With this patch, that
is not required.
2. Typically, model number has to be added to the white list when the
system is under development. With this change, that is not required.
Signed-off-by: Jordan Hargrave <jordan_hargrave@dell.com>
Signed-off-by: Narendra K <narendra_k@dell.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
The Linux kernel assigns BARs that a BIOS did not assign, most likely
to handle broken BIOSes that didn't enumerate the devices correctly.
On UV the BIOS purposely doesn't assign I/O BARs for certain devices/
drivers we know don't use them (examples, LSI SAS, Qlogic FC, ...).
We purposely don't assign these I/O BARs because I/O Space is a very
limited resource. There is only 64k of I/O Space, and in a PCIe
topology that space gets divided up into 4k chucks (this is due to
the fact that a pci-to-pci bridge's I/O decoder is aligned at 4k)...
Thus a system can have at most 16 cards with I/O BARs: (64k / 4k = 16)
SGI needs to scale to >16 devices with I/O BARs. So by not assigning
I/O BARs on devices we know don't use them, we can do that (iff the
kernel doesn't go and assign these BARs that the BIOS purposely didn't
assign).
This patch will not assign a resource to a device BAR if that BAR was
not assigned by the BIOS, and the kernel cmdline option 'pci=nobar'
was specified. This patch is closely modeled after the 'pci=norom'
option that currently exists in the tree.
Signed-off-by: Mike Habeck <habeck@sgi.com>
Signed-off-by: Mike Travis <travis@sgi.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
pci_config_lock must be a real spinlock in preempt-rt. Convert it to
raw_spinlock. No change for !RT kernels.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
percpu.h is included by sched.h and module.h and thus ends up being
included when building most .c files. percpu.h includes slab.h which
in turn includes gfp.h making everything defined by the two files
universally available and complicating inclusion dependencies.
percpu.h -> slab.h dependency is about to be removed. Prepare for
this change by updating users of gfp and slab facilities include those
headers directly instead of assuming availability. As this conversion
needs to touch large number of source files, the following script is
used as the basis of conversion.
http://userweb.kernel.org/~tj/misc/slabh-sweep.py
The script does the followings.
* Scan files for gfp and slab usages and update includes such that
only the necessary includes are there. ie. if only gfp is used,
gfp.h, if slab is used, slab.h.
* When the script inserts a new include, it looks at the include
blocks and try to put the new include such that its order conforms
to its surrounding. It's put in the include block which contains
core kernel includes, in the same order that the rest are ordered -
alphabetical, Christmas tree, rev-Xmas-tree or at the end if there
doesn't seem to be any matching order.
* If the script can't find a place to put a new include (mostly
because the file doesn't have fitting include block), it prints out
an error message indicating which .h file needs to be added to the
file.
The conversion was done in the following steps.
1. The initial automatic conversion of all .c files updated slightly
over 4000 files, deleting around 700 includes and adding ~480 gfp.h
and ~3000 slab.h inclusions. The script emitted errors for ~400
files.
2. Each error was manually checked. Some didn't need the inclusion,
some needed manual addition while adding it to implementation .h or
embedding .c file was more appropriate for others. This step added
inclusions to around 150 files.
3. The script was run again and the output was compared to the edits
from #2 to make sure no file was left behind.
4. Several build tests were done and a couple of problems were fixed.
e.g. lib/decompress_*.c used malloc/free() wrappers around slab
APIs requiring slab.h to be added manually.
5. The script was run on all .h files but without automatically
editing them as sprinkling gfp.h and slab.h inclusions around .h
files could easily lead to inclusion dependency hell. Most gfp.h
inclusion directives were ignored as stuff from gfp.h was usually
wildly available and often used in preprocessor macros. Each
slab.h inclusion directive was examined and added manually as
necessary.
6. percpu.h was updated not to include slab.h.
7. Build test were done on the following configurations and failures
were fixed. CONFIG_GCOV_KERNEL was turned off for all tests (as my
distributed build env didn't work with gcov compiles) and a few
more options had to be turned off depending on archs to make things
build (like ipr on powerpc/64 which failed due to missing writeq).
* x86 and x86_64 UP and SMP allmodconfig and a custom test config.
* powerpc and powerpc64 SMP allmodconfig
* sparc and sparc64 SMP allmodconfig
* ia64 SMP allmodconfig
* s390 SMP allmodconfig
* alpha SMP allmodconfig
* um on x86_64 SMP allmodconfig
8. percpu.h modifications were reverted so that it could be applied as
a separate patch and serve as bisection point.
Given the fact that I had only a couple of failures from tests on step
6, I'm fairly confident about the coverage of this conversion patch.
If there is a breakage, it's likely to be something in one of the arch
headers which should be easily discoverable easily on most builds of
the specific arch.
Signed-off-by: Tejun Heo <tj@kernel.org>
Guess-its-ok-by: Christoph Lameter <cl@linux-foundation.org>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Lee Schermerhorn <Lee.Schermerhorn@hp.com>
The PCI initialization in pci_subsys_init() is a mess. pci_numaq_init,
pci_acpi_init, pci_visws_init and pci_legacy_init are called and each
implementation checks and eventually modifies the global variable
pcibios_scanned.
x86_init functions allow us to do this more elegant. The pci.init
function pointer is preset to pci_legacy_init. numaq, acpi and visws
can modify the pointer in their early setup functions. The functions
return 0 when they did the full initialization including bus scan. A
non zero return value indicates that pci_legacy_init needs to be
called either because the selected function failed or wants the
generic bus scan in pci_legacy_init to happen (e.g. visws).
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
LKML-Reference: <43F901BD926A4E43B106BF17856F07559FB80CFE@orsmsx508.amr.corp.intel.com>
Acked-by: Jesse Barnes <jbarnes@virtuousgeek.org>
Signed-off-by: Jacob Pan <jacob.jun.pan@intel.com>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Instead of the PCI code needing to have code to determine the
cacheline size of each processor, use the data the cpu identification
code should have already determined during early boot.
(The vendor checks are also incomplete, and don't take into account
modern CPUs)
I've been carrying a variant of this code in Fedora for a while,
that prints debug information. There are a number of cases where we
are currently setting the PCI cacheline size to 32 bytes, when the CPU
cacheline size is 64 bytes. With this patch, we set them both the same.
Signed-off-by: Dave Jones <davej@redhat.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
Till now, CLS has been determined either by arch code or as
L1_CACHE_BYTES. Only x86 and ia64 set CLS explicitly and x86 doesn't
always get it right. On most configurations, the chance is that
firmware configures the correct value during boot.
This patch makes pci_init() determine CLS by looking at what firmware
has configured. It scans all devices and if all non-zero values
agree, the value is used. If none is configured or there is a
disagreement, pci_dfl_cache_line_size is used. arch can set the dfl
value (via PCI_CACHE_LINE_BYTES or pci_dfl_cache_line_size) or
override the actual one.
ia64, x86 and sparc64 updated to set the default cls instead of the
actual one.
While at it, declare pci_cache_line_size and pci_dfl_cache_line_size
in pci.h and drop private declarations from arch code.
Signed-off-by: Tejun Heo <tj@kernel.org>
Acked-by: David Miller <davem@davemloft.net>
Acked-by: Greg KH <gregkh@suse.de>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Tony Luck <tony.luck@intel.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
The current mp_bus_to_node array is initialized only by AMD specific
code, since AMD platforms have registers that can be used for
determining mode numbers. On new Intel platforms it's necessary to
initialize this array as well though, otherwise all PCI node numbers
will be 0, when in fact they should be -1 (indicating that I/O isn't
tied to any particular node).
So move the mp_bus_to_node code into the common PCI code, and
initialize it early with a default value of -1. This may be overridden
later by arch code (e.g. the AMD code).
With this change, PCI consistent memory and other node specific
allocations (e.g. skbuff allocs) should occur on the "current" node.
If, for performance reasons, applications want to be bound to specific
nodes, they should open their devices only after being pinned to the
CPU where they'll run, for maximum locality.
Acked-by: Yinghai Lu <yinghai@kernel.org>
Tested-by: Jesse Brandeburg <jesse.brandeburg@gmail.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
This reverts commit 9e9f46c44e.
Quoting from the commit message:
"At this point, it seems to solve more problems than it causes, so let's
try using it by default. It's an easy revert if it ends up causing
trouble."
And guess what? The _CRS code causes trouble.
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
At this point, it seems to solve more problems than it causes, so let's try using it by default. It's an easy revert if it ends up causing trouble.
Reviewed-by: Yinghai Lu <yhlu.kernel@gmail.com>
Acked-by: Bjorn Helgaas <bjorn.helgaas@hp.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
Rename set_pci_bus_resources_arch_default to x86_pci_root_bus_res_quirks, move
the weak version from common.c to i386.c, and before calling, make sure it's a
root bus.
Reviewed-by: Matthew Wilcox <willy@linux.intel.com>
Signed-off-by: Yinghai Lu <yinghai@kernel.org>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
Commit 30a18d6c3f introduced a new
function to set the PCI bus resources. Unfortunately, neither the
author, nor the committers seemed to know that we already have somewhere
to do that -- pcibios_fixup_bus(). This patch moves the hook (used only
by the K8 code) into x86-specific code where it should have been in the
first place.
Cc: Yinghai Lu <yinghai.lu@sun.com>
Signed-off-by: Matthew Wilcox <willy@linux.intel.com>
Acked-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
Impact: debuggability and micro-optimization
Putting whatever is possible into the (final) .rodata section increases
the likelihood of catching memory corruption bugs early, and reduces
false cache line sharing.
Signed-off-by: Jan Beulich <jbeulich@novell.com>
LKML-Reference: <49B909A5.76E4.0078.0@novell.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
pcibios_enable_device() and pcibios_disable_device() don't handle
IRQs for devices that have MSI enabled and it should treat the
devices with MSI-X enabled in the same way.
Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
Acked-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
Bjorn Helgaas
Yinghai Lu
Jan Beulich
Narendra_K@Dell.com
Alex Nixon
Mike Habeck
Thomas Gleixner
Tejun Heo
Linus Torvalds
Bjorn Helgaas
Dave Jones
Jesse Barnes
Matthew Wilcox
Jan Beulich
Rafael J. Wysocki