linux-toradex.git/arch/x86/pci, branch v3.1

x86/PCI: use host bridge _CRS info on ASUS M2V-MX SE

2011-10-06T23:10:37+00:00

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 2491762cfb47 ("x86/PCI: use host bridge _CRS info on
ASRock ALiveSATA2-GLAN") whose commit message should be read for further
information.

Since commit 3e3da00c01d0 ("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: [] 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:[]  [] 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:
     [] ? __pm_runtime_set_status+0x162/0x186
     [] ? local_pci_probe+0x49/0x92
     [] ? do_work_for_cpu+0x0/0x1b
     [] ? do_work_for_cpu+0x0/0x1b
     [] ? do_work_for_cpu+0xb/0x1b
     [] ? kthread+0x7a/0x82
     [] ? kernel_thread_helper+0x4/0x10
     [] ? kthread+0x0/0x82
     [] ? 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  [] azx_probe+0x3ad/0x86b [snd_hda_intel]
     RSP 
    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 
Cc: Ingo Molnar 
Cc: H. Peter Anvin 
Cc: x86@kernel.org
Signed-off-by: Paul Menzel 
Signed-off-by: Bjorn Helgaas 
Acked-by: Jesse Barnes 
Signed-off-by: Linus Torvalds

Fix pointer dereference before call to pcie_bus_configure_settings

2011-09-10T02:49:58+00:00

Commit b03e7495a862 ("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 
Signed-off-by: Shyam Iyer 
Signed-off-by: Jon Mason 
Acked-by: Jesse Barnes 
Signed-off-by: Linus Torvalds

PCI: Set PCI-E Max Payload Size on fabric

2011-08-01T18:49:16+00:00

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 
Signed-off-by: Jesse Barnes

Merge branch 'linux-next' of git://git.kernel.org/pub/scm/linux/kernel/git/jbarnes/pci-2.6

2011-07-30T06:35:05+00:00

* 'linux-next' of git://git.kernel.org/pub/scm/linux/kernel/git/jbarnes/pci-2.6:
  PCI: remove printks about disabled bridge windows
  PCI: fold pci_calc_resource_flags() into decode_bar()
  PCI: treat mem BAR type "11" (reserved) as 32-bit, not 64-bit, BAR
  PCI: correct pcie_set_readrq write size
  PCI: pciehp: change wait time for valid configuration access
  x86/PCI: Preserve existing pci=bfsort whitelist for Dell systems
  PCI: ARI is a PCIe v2 feature
  x86/PCI: quirks: Use pci_dev->revision
  PCI: Make the struct pci_dev * argument of pci_fixup_irqs const.
  PCI hotplug: cpqphp: use pci_dev->vendor
  PCI hotplug: cpqphp: use pci_dev->subsystem_{vendor|device}
  x86/PCI: config space accessor functions should not ignore the segment argument
  PCI: Assign values to 'pci_obff_signal_type' enumeration constants
  x86/PCI: reduce severity of host bridge window conflict warnings
  PCI: enumerate the PCI device only removed out PCI hieratchy of OS when re-scanning PCI
  PCI: PCIe AER: add aer_recover_queue
  x86/PCI: select direct access mode for mmconfig option
  PCI hotplug: Rename is_ejectable which also exists in dock.c

Merge branch 'x86-uv-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip

2011-07-23T00:04:55+00:00

* 'x86-uv-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip:
  x86, UV: Correct UV2 BAU destination timeout
  x86, UV: Correct failed topology memory leak
  x86, UV: Remove cpumask_t from the stack
  x86, UV: Rename hubmask to pnmask
  x86, UV: Correct reset_with_ipi()
  x86, UV: Allow for non-consecutive sockets
  x86, UV: Inline header file functions
  x86, UV: Fix smp_processor_id() use in a preemptable region
  x66, UV: Enable 64-bit ACPI MFCG support for SGI UV2 platform
  x86, UV: Clean up uv_mmrs.h

x86/PCI: Preserve existing pci=bfsort whitelist for Dell systems

2011-07-22T15:44:28+00:00

Commit 6e8af08dfa40b747002207d3ce8e8b43a050d99f 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 
Signed-off-by: Narendra K 
Signed-off-by: Jesse Barnes

PCI: Make the struct pci_dev * argument of pci_fixup_irqs const.

2011-07-22T15:26:06+00:00

Aside of the usual motivation for constification,  this function has a
history of being abused a hook for interrupt and other fixups so I turned
this function const ages ago in the MIPS code but it should be done
treewide.

Due to function pointer passing in varous places a few other functions
had to be constified as well.

Signed-off-by: Ralf Baechle 
To: Anton Vorontsov 
To: Chris Metcalf 
To: Colin Cross 
Acked-by: "David S. Miller" 
To: Eric Miao 
To: Erik Gilling 
Acked-by: Guan Xuetao 
To: "H. Peter Anvin" 
To: Imre Kaloz 
To: Ingo Molnar 
To: Ivan Kokshaysky 
To: Jesse Barnes 
To: Krzysztof Halasa 
To: Lennert Buytenhek 
To: Matt Turner 
To: Nicolas Pitre 
To: Olof Johansson 
Acked-by: Paul Mundt 
To: Richard Henderson 
To: Russell King 
To: Thomas Gleixner 
Cc: Andrew Morton 
Cc: linux-alpha@vger.kernel.org
Cc: linux-arm-kernel@lists.infradead.org
Cc: linux-kernel@vger.kernel.org
Cc: linux-mips@linux-mips.org
Cc: linux-pci@vger.kernel.org
Cc: linux-sh@vger.kernel.org
Cc: linux-tegra@vger.kernel.org
Cc: sparclinux@vger.kernel.org
Cc: x86@kernel.org
Signed-off-by: Jesse Barnes

x86/PCI: config space accessor functions should not ignore the segment argument

2011-07-22T15:25:41+00:00

Without this change, the majority of the raw PCI config space access
functions silently ignore a non-zero segment argument, which is
certainly wrong.

Apart from pci_direct_conf1, all other non-MMCFG access methods get
used only for non-extended accesses (i.e. assigned to raw_pci_ops
only). Consequently, with the way raw_pci_{read,write}() work, it would
be a coding error to call these functions with a non-zero segment (with
the current call flow this cannot happen afaict).

The access method 1 accessor, as it can be used for extended accesses
(on AMD systems) instead gets checks added for the passed in segment to
be zero. This would be the case when on such a system having multiple
PCI segments (don't know whether any exist in practice) MMCFG for some
reason is not usable, and method 1 gets selected for doing extended
accesses. Rather than accessing the wrong device's config space, the
function will now error out.

v2: Convert BUG_ON() to WARN_ON(), and extend description as per Ingo's
request.

Signed-off-by: Jan Beulich 
Reviewed-by: Ingo Molnar 
Signed-off-by: Jesse Barnes

x86/PCI: reduce severity of host bridge window conflict warnings

2011-07-22T15:25:39+00:00

Host bridge windows are top-level resources, so if we find a host bridge
window conflict, it's probably with a hard-coded legacy reservation.
Moving host bridge windows is theoretically possible, but we don't support
it; we just ignore windows with conflicts, and it's not worth making this
a user-visible error.

Reported-and-tested-by: Jools Wills 
References: https://bugzilla.kernel.org/show_bug.cgi?id=38522
Reported-by: Das 
References: https://bugzilla.kernel.org/show_bug.cgi?id=16497
Signed-off-by: Bjorn Helgaas 
Signed-off-by: Jesse Barnes

xen/pci: Use 'acpi_gsi_to_irq' value unconditionally.

2011-07-11T17:19:34+00:00

In the past we would only use the function's value if the
returned value was not equal to 'acpi_sci_override_gsi'. Meaning
that the INT_SRV_OVR values for global and source irq were different.
But it is OK to use the function's value even when the global
and source irq are the same.

Signed-off-by: Konrad Rzeszutek Wilk