- To avoid bus errors, enable PASID only if entire path supports End-End
TLP prefixes (Sinan Kaya)
- Unify slot and bus reset functions and remove hotplug knowledge from
callers (Sinan Kaya)
- Add Function-Level Reset quirks for Intel and Samsung NVMe devices to
fix guest reboot issues (Alex Williamson)
- Add function 1 DMA alias quirk for Marvell 88SS9183 PCIe SSD Controller
(Bjorn Helgaas)
* pci/virtualization:
PCI: Add function 1 DMA alias quirk for Marvell 88SS9183
PCI: Delay after FLR of Intel DC P3700 NVMe
PCI: Disable Samsung SM961/PM961 NVMe before FLR
PCI: Export pcie_has_flr()
PCI: Rename pci_try_reset_bus() to pci_reset_bus()
PCI: Deprecate pci_reset_bus() and pci_reset_slot() functions
PCI: Unify try slot and bus reset API
PCI: Hide pci_reset_bridge_secondary_bus() from drivers
IB/hfi1: Use pci_try_reset_bus() for initiating PCI Secondary Bus Reset
PCI: Handle error return from pci_reset_bridge_secondary_bus()
PCI/IOV: Tidy pci_sriov_set_totalvfs()
PCI: Enable PASID only if entire path supports End-End TLP prefixes
# Conflicts:
# drivers/pci/hotplug/pciehp_hpc.c
- Add DMA alias quirk for Microsemi Switchtec NTB (Doug Meyer)
- Expand documentation for pci_add_dma_alias() (Logan Gunthorpe)
* pci/switchtec:
PCI: Expand documentation for pci_add_dma_alias()
PCI: Add DMA alias quirk for Microsemi Switchtec NTB
switchtec: Use generic PCI Vendor ID and Class Code
# Conflicts:
# drivers/pci/quirks.c
- Clean up devm_of_pci_get_host_bridge_resources() resource allocation
(Jan Kiszka)
- Fixup resizable BARs after suspend/resume (Christian König)
- Make "pci=earlydump" generic (Sinan Kaya)
- Fix ROM BAR access routines to stay in bounds and check for signature
correctly (Rex Zhu)
* pci/resource:
PCI: Make pci_get_rom_size() static
PCI: Add check code for last image indicator not set
PCI: Avoid accessing memory outside the ROM BAR
PCI: Make early dump functionality generic
PCI: Cleanup PCI_REBAR_CTRL_BAR_SHIFT handling
PCI: Restore resized BAR state on resume
PCI: Clean up resource allocation in devm_of_pci_get_host_bridge_resources()
# Conflicts:
# Documentation/admin-guide/kernel-parameters.txt
- Add "pci=disable_acs_redir=" parameter to disable ACS redirection for
peer-to-peer DMA support (we don't have the peer-to-peer support yet;
this is just one piece) (Logan Gunthorpe)
* pci/peer-to-peer:
PCI: Add ACS Redirect disable quirk for Intel Sunrise Point
PCI: Add device-specific ACS Redirect disable infrastructure
PCI: Convert device-specific ACS quirks from NULL termination to ARRAY_SIZE
PCI: Add "pci=disable_acs_redir=" parameter for peer-to-peer support
PCI: Allow specifying devices using a base bus and path of devfns
PCI: Make specifying PCI devices in kernel parameters reusable
PCI: Hide ACS quirk declarations inside PCI core
- Mark fall-through switch cases before enabling -Wimplicit-fallthrough
(Gustavo A. R. Silva)
- Move DMA-debug PCI init from arch code to PCI core (Christoph Hellwig)
- Fix pci_request_irq() usage of IRQF_ONESHOT when no handler is supplied
(Heiner Kallweit)
- Unify PCI and DMA direction #defines (Shunyong Yang)
- Add PCI_DEVICE_DATA() macro (Andy Shevchenko)
- Check for VPD completion before checking for timeout (Bert Kenward)
- Limit Netronome NFP5000 config space size to work around erratum (Jakub
Kicinski)
* pci/misc:
PCI: Limit config space size for Netronome NFP5000
PCI/VPD: Check for VPD access completion before checking for timeout
PCI: Add PCI_DEVICE_DATA() macro to fully describe device ID entry
PCI: Unify PCI and normal DMA direction definitions
PCI: Use IRQF_ONESHOT if pci_request_irq() called with no handler
PCI: Call dma_debug_add_bus() for pci_bus_type from PCI core
PCI: Mark fall-through switch cases before enabling -Wimplicit-fallthrough
# Conflicts:
# drivers/pci/hotplug/pciehp_ctrl.c
- Work around IDT switch ACS Source Validation erratum (James
Puthukattukaran)
- Emit diagnostics for all cases of PCIe Link downtraining (Links
operating slower than they're capable of) (Alexandru Gagniuc)
- Skip VFs when configuring Max Payload Size (Myron Stowe)
- Reduce Root Port Max Payload Size if necessary when hot-adding a device
below it (Myron Stowe)
* pci/enumeration:
PCI: Match Root Port's MPS to endpoint's MPSS as necessary
PCI: Skip MPS logic for Virtual Functions (VFs)
PCI: Check for PCIe Link downtraining
PCI: Workaround IDT switch ACS Source Validation erratum
- Defer DPC event handling to work queue (Keith Busch)
- Use threaded IRQ for DPC bottom half (Keith Busch)
- Print AER status while handling DPC events (Keith Busch)
* pci/dpc:
PCI/DPC: Remove indirection waiting for inactive link
PCI/DPC: Use threaded IRQ for bottom half handling
PCI/DPC: Print AER status in DPC event handling
PCI/DPC: Remove rp_pio_status from dpc struct
PCI/DPC: Defer event handling to work queue
PCI/DPC: Leave interrupts enabled while handling event
- Use sysfs_match_string() to simplify ASPM sysfs parsing (Andy
Shevchenko)
- Remove unnecessary includes of <linux/pci-aspm.h> (Bjorn Helgaas)
* pci/aspm:
PCI: Remove unnecessary include of <linux/pci-aspm.h>
iwlwifi: Remove unnecessary include of <linux/pci-aspm.h>
ath9k: Remove unnecessary include of <linux/pci-aspm.h>
igb: Remove unnecessary include of <linux/pci-aspm.h>
PCI/ASPM: Convert to use sysfs_match_string() helper
- Decode AER errors with names similar to "lspci" (Tyler Baicar)
- Expose AER statistics in sysfs (Rajat Jain)
- Clear AER status bits selectively based on the type of recovery (Oza
Pawandeep)
- Honor "pcie_ports=native" even if HEST sets FIRMWARE_FIRST (Alexandru
Gagniuc)
- Don't clear AER status bits if we're using the "Firmware-First"
strategy where firmware owns the registers (Alexandru Gagniuc)
* pci/aer:
PCI/AER: Don't clear AER bits if error handling is Firmware-First
PCI/AER: Remove duplicate PCI_EXP_AER_FLAGS definition
PCI/portdrv: Remove pcie_portdrv_err_handler.slot_reset
PCI/AER: Clear device status bits during ERR_COR handling
PCI/AER: Clear device status bits during ERR_FATAL and ERR_NONFATAL
PCI/AER: Remove ERR_FATAL code from ERR_NONFATAL path
PCI/AER: Factor out ERR_NONFATAL status bit clearing
PCI/AER: Clear only ERR_NONFATAL bits during non-fatal recovery
PCI/AER: Clear only ERR_FATAL status bits during fatal recovery
PCI/AER: Honor "pcie_ports=native" even if HEST sets FIRMWARE_FIRST
PCI/AER: Add sysfs attributes for rootport cumulative stats
PCI/AER: Add sysfs attributes to provide AER stats and breakdown
PCI/AER: Define aer_stats structure for AER capable devices
PCI/AER: Move internal declarations to drivers/pci/pci.h
PCI/AER: Adopt lspci names for AER error decoding
PCI/AER: Expose internal API for obtaining AER information
# Conflicts:
# drivers/pci/pci.h
If the platform requests Firmware-First error handling, firmware is
responsible for reading and clearing AER status bits. If OSPM also clears
them, we may miss errors. See ACPI v6.2, sec 18.3.2.5 and 18.4.
This race is mostly of theoretical significance, as it is not easy to
reasonably demonstrate it in testing.
Signed-off-by: Alexandru Gagniuc <mr.nuke.me@gmail.com>
[bhelgaas: add similar guards to pci_cleanup_aer_uncorrect_error_status()
and pci_aer_clear_fatal_status()]
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Like the NFP4000 and NFP6000, the NFP5000 as an erratum where reading/
writing to PCI config space addresses above 0x600 can cause the NFP to
generate PCIe completion timeouts.
Limit the NFP5000's PF's config space size to 0x600 bytes as is already
done for the NFP4000 and NFP6000.
The NFP5000's VF is 0x6003 (PCI_DEVICE_ID_NETRONOME_NFP6000_VF), the same
device ID as the NFP6000's VF. Thus, its config space is already limited
by the existing use of quirk_nfp6000().
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Reviewed-by: Tony Egan <tony.egan@netronome.com>
If flag IRQCHIP_ONESHOT_SAFE isn't set for an irqchip and we have a
threaded interrupt with no primary handler, flag IRQF_ONESHOT needs to be
set for the interrupt, causing some overhead in the threaded interrupt
handler. For more detailed explanation also check following comment in
__setup_irq():
The interrupt was requested with handler = NULL, so we use the default
primary handler for it. But it does not have the oneshot flag set. In
combination with level interrupts this is deadly, because the default
primary handler just wakes the thread, then the irq lines is reenabled,
but the device still has the level irq asserted. Rinse and repeat....
While this works for edge type interrupts, we play it safe and reject
unconditionally because we can't say for sure which type this interrupt
really has. The type flags are unreliable as the underlying chip
implementation can override them.
Another comment in __setup_irq() gives a hint already that this
overhead can be avoided for PCI-MSI:
Some irq chips like MSI based interrupts are per se one shot safe. Check
the chip flags, so we can avoid the unmask dance at the end of the
threaded handler for those.
Following this let's mark all PCI-MSI irqchips as oneshot-safe.
See also discussion here:
https://lkml.kernel.org/r/alpine.DEB.2.21.1808032136490.1658@nanos.tec.linutronix.de
Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Previously we checked the timeout before checking the VPD access completion
bit. On a very heavily loaded system this can cause VPD access to timeout.
Check the completion bit before checking the timeout.
Signed-off-by: Bert Kenward <bkenward@solarflare.com>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
There are a lot of examples in the kernel where PCI_VDEVICE() is used and
still looks not so convenient due to additional driver_data field attached.
Introduce PCI_DEVICE_DATA() macro to fully describe device ID entry in
shortest possible form. For example,
before:
{ PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_MRFLD),
(kernel_ulong_t) &dwc3_pci_mrfld_properties, },
after:
{ PCI_DEVICE_DATA(INTEL, MRFLD, &dwc3_pci_mrfld_properties) },
Drivers can be converted later on in independent way.
While here, remove the unused macro with the same name from Ralink wireless
driver.
Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Acked-by: Kalle Valo <kvalo@codeaurora.org> # for rt2x00
In commit 27d868b5e6 ("PCI: Set MPS to match upstream bridge"), we made
sure every device's MPS setting matches its upstream bridge, making it more
likely that a hot-added device will work in a system with an optimized MPS
configuration.
Recently I've started encountering systems where the endpoint device's MPSS
capability is less than its Root Port's current MPS value, thus the
endpoint is not capable of matching its upstream bridge's MPS setting (see:
bugzilla via "Link:" below). This leaves the system vulnerable - the
upstream Root Port could respond with larger TLPs than the device can
handle, and the device will consider them to be 'Malformed'.
One could use the "pci=pcie_bus_safe" kernel parameter to work around the
issue, but that forces a user to supply a kernel parameter to get the
system to function reliably and may end up limiting MPS settings of other
unrelated, sub-topologies which could benefit from maintaining their larger
values.
Augment Keith's approach to include tuning down a Root Port's MPS setting
when its hot-added endpoint device is not capable of matching it.
Link: https://bugzilla.kernel.org/show_bug.cgi?id=200527
Signed-off-by: Myron Stowe <myron.stowe@redhat.com>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Acked-by: Jon Mason <jdmason@kudzu.us>
Cc: Keith Busch <keith.busch@intel.com>
Cc: Sinan Kaya <okaya@kernel.org>
Cc: Dongdong Liu <liudongdong3@huawei.com>
PCIe r4.0, sec 9.3.5.4, "Device Control Register", shows both
Max_Payload_Size (MPS) and Max_Read_request_Size (MRRS) to be 'RsvdP' for
VFs. Just prior to the table it states:
"PF and VF functionality is defined in Section 7.5.3.4 except where
noted in Table 9-16. For VF fields marked 'RsvdP', the PF setting
applies to the VF."
All of which implies that with respect to Max_Payload_Size Supported
(MPSS), MPS, and MRRS values, we should not be paying any attention to the
VF's fields, but rather only to the PF's. Only looking at the PF's fields
also logically makes sense as it's the sole physical interface to the PCIe
bus.
Link: https://bugzilla.kernel.org/show_bug.cgi?id=200527
Fixes: 27d868b5e6 ("PCI: Set MPS to match upstream bridge")
Signed-off-by: Myron Stowe <myron.stowe@redhat.com>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Cc: stable@vger.kernel.org # 4.3+
Cc: Keith Busch <keith.busch@intel.com>
Cc: Sinan Kaya <okaya@kernel.org>
Cc: Dongdong Liu <liudongdong3@huawei.com>
Cc: Jon Mason <jdmason@kudzu.us>
When both ends of a PCIe Link are capable of a higher bandwidth than is
currently in use, the Link is said to be "downtrained". A downtrained Link
may indicate hardware or configuration problems in the system, but it's
hard to identify such Links from userspace.
Refactor pcie_print_link_status() so it continues to always print PCIe
bandwidth information, as several NIC drivers desire.
Add a new internal __pcie_print_link_status() to emit a message only when a
device's bandwidth is constrained by the fabric and call it from the PCI
core for all devices, which identifies all downtrained Links. It also
emits messages for a few cases that are technically not downtrained, such
as a x4 device in an open-ended x1 slot.
Signed-off-by: Alexandru Gagniuc <mr.nuke.me@gmail.com>
[bhelgaas: changelog, move __pcie_print_link_status() declaration to
drivers/pci/, rename pcie_check_upstream_link() to
pcie_report_downtraining()]
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Intel Sunrise Point PCH hardware has an implementation of the ACS bits that
does not comply with the PCIe standard. Add a device-specific quirk,
pci_quirk_disable_intel_spt_pch_acs_redir() to disable ACS Redirection on
this system.
Signed-off-by: Logan Gunthorpe <logang@deltatee.com>
[bhelgaas: changelog, split to separate patch]
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Reviewed-by: Alex Williamson <alex.williamson@redhat.com>
Intel Sunrise Point (SPT) PCH hardware has an implementation of the ACS
bits that does not comply with the PCIe standard. To deal with this we
need device-specific quirks to disable ACS redirection.
Add a new pci_dev_specific_disable_acs_redir() quirk and a new
.disable_acs_redir() function pointer for use by non-compliant devices. No
functional change intended.
Signed-off-by: Logan Gunthorpe <logang@deltatee.com>
[bhelgaas: split to separate patch, move
pci_dev_specific_disable_acs_redir() declarations to drivers/pci/pci.h]
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Reviewed-by: Alex Williamson <alex.williamson@redhat.com>
Convert the search for device-specific ACS enable quirks from searching a
NULL-terminated array to iterating through the array, which is always
fixed-size anyway. No functional change intended.
Signed-off-by: Logan Gunthorpe <logang@deltatee.com>
[bhelgaas: changelog, split to separate patch for reviewability]
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Reviewed-by: Alex Williamson <alex.williamson@redhat.com>
To support peer-to-peer traffic on a segment of the PCI hierarchy, we must
disable the ACS redirect bits for select PCI bridges. The bridges must be
selected before the devices are discovered by the kernel and the IOMMU
groups created. Therefore, add a kernel command line parameter to specify
devices which must have their ACS bits disabled.
The new parameter takes a list of devices separated by a semicolon. Each
device specified will have its ACS redirect bits disabled. This is
similar to the existing 'resource_alignment' parameter.
The ACS Request P2P Request Redirect, P2P Completion Redirect and P2P
Egress Control bits are disabled, which is sufficient to always allow
passing P2P traffic uninterrupted. The bits are set after the kernel
(optionally) enables the ACS bits itself. It is also done regardless of
whether the kernel or platform firmware sets the bits.
If the user tries to disable the ACS redirect for a device without the ACS
capability, print a warning to dmesg.
Signed-off-by: Logan Gunthorpe <logang@deltatee.com>
[bhelgaas: reorder to add the generic code first and move the
device-specific quirk to subsequent patches]
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Reviewed-by: Stephen Bates <sbates@raithlin.com>
Reviewed-by: Alex Williamson <alex.williamson@redhat.com>
Acked-by: Christian König <christian.koenig@amd.com>
When specifying PCI devices on the kernel command line using a
bus/device/function address, bus numbers can change when adding or
replacing a device, changing motherboard firmware, or applying kernel
parameters like "pci=assign-buses". When bus numbers change, it's likely
the command line tweak will be applied to the wrong device.
Therefore, it is useful to be able to specify devices with a base bus
number and the path of devfns needed to get to it, similar to the "device
scope" structure in the Intel VT-d spec, Section 8.3.1.
Thus, we add an option to specify devices in the following format:
[<domain>:]<bus>:<device>.<func>[/<device>.<func>]*
The path can be any segment within the PCI hierarchy of any length and
determined through the use of 'lspci -t'. When specified this way, it is
less likely that a renumbered bus will result in a valid device
specification and the tweak won't be applied to the wrong device.
Signed-off-by: Logan Gunthorpe <logang@deltatee.com>
[bhelgaas: use "device" instead of "slot" in documentation since that's the
usual language in the PCI specs]
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Reviewed-by: Stephen Bates <sbates@raithlin.com>
Reviewed-by: Alex Williamson <alex.williamson@redhat.com>
Acked-by: Christian König <christian.koenig@amd.com>
Separate out the code to match a PCI device with a string (typically
originating from a kernel parameter) from the
pci_specified_resource_alignment() function into its own helper function.
While we are at it, this change fixes the kernel style of the function
(fixing a number of long lines and extra parentheses).
Additionally, make the analogous change to the kernel parameter
documentation: Separate the description of how to specify a PCI device
into its own section at the head of the "pci=" parameter.
This patch should have no functional alterations.
Signed-off-by: Logan Gunthorpe <logang@deltatee.com>
[bhelgaas: use "device" instead of "slot" in documentation since that's the
usual language in the PCI specs]
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Reviewed-by: Stephen Bates <sbates@raithlin.com>
Reviewed-by: Alex Williamson <alex.williamson@redhat.com>
Acked-by: Christian König <christian.koenig@amd.com>
Move declarations for these functions:
pci_dev_specific_acs_enabled()
pci_dev_specific_enable_acs()
from include/linux/pci.h to drivers/pci/pci.h because nothing outside the
PCI core needs to use them.
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Add a device-specific reset for Intel DC P3700 NVMe device which exhibits a
timeout failure in drivers waiting for the ready status to update after
NVMe enable if the driver interacts with the device too soon after FLR. As
this has been observed in device assignment scenarios, resolve this with a
device-specific reset quirk to add an additional, heuristically determined,
delay after the FLR completes.
Link: https://bugzilla.redhat.com/show_bug.cgi?id=1592654
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
The Samsung SM961/PM961 (960 EVO) sometimes fails to return from FLR with
the PCI config space reading back as -1. A reproducible instance of this
behavior is resolved by clearing the enable bit in the NVMe configuration
register and waiting for the ready status to clear (disabling the NVMe
controller) prior to FLR.
Link: https://bugzilla.redhat.com/show_bug.cgi?id=1542494
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
pcie_flr() suggests pcie_has_flr() to ensure that PCIe FLR support is
present prior to calling. pcie_flr() is exported while pcie_has_flr()
is not. Resolve this.
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Several PCI core files include pci-aspm.h even though they don't need
anything provided by that file. Remove the unnecessary includes of it.
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Reviewed-by: Sinan Kaya <okaya@kernel.org>
This part of the iwlwifi driver doesn't need anything provided by
pci-aspm.h, so remove the unnecessary include of it.
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Reviewed-by: Sinan Kaya <okaya@kernel.org>
Acked-by: Luca Coelho <luciano.coelho@intel.com>
The ath9k driver doesn't need anything provided by pci-aspm.h, so remove
the unnecessary include of it.
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Reviewed-by: Sinan Kaya <okaya@kernel.org>
Acked-by: Kalle Valo <kvalo@codeaurora.org>
The igb driver doesn't need anything provided by pci-aspm.h, so remove
the unnecessary include of it.
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Reviewed-by: Sinan Kaya <okaya@kernel.org>
Acked-by: Alexander Duyck <alexander.h.duyck@intel.com>
Acked-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
The sysfs_match_string() helper returns index of the matching string in an
array. Use it in pcie_aspm_set_policy() to simplify the code.
Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
[bhelgaas: squash sysfs_match_string() fix into original patch for issue
Reported-by: Heiner Kallweit <hkallweit1@gmail.com>]
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Current DMA direction definitions in pci-dma-compat.h and dma-direction.h
are mirrored in value. Unify them to enhance readability and avoid
possible inconsistency.
Signed-off-by: Shunyong Yang <shunyong.yang@hxt-semitech.com>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Cc: Joey Zheng <yu.zheng@hxt-semitech.com>
PCI_EXP_AER_FLAGS was defined twice (with identical definitions), once
under #ifdef CONFIG_ACPI_APEI, and again at the top level. This looks like
my merge error from these commits:
fd3362cb73 ("PCI/AER: Squash aerdrv_core.c into aerdrv.c")
41cbc9eb1a ("PCI/AER: Squash ecrc.c into aerdrv.c")
Remove the duplicate PCI_EXP_AER_FLAGS definition.
Fixes: 41cbc9eb1a ("PCI/AER: Squash ecrc.c into aerdrv.c")
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Reviewed-by: Oza Pawandeep <poza@codeaurora.org>
On driver probe and on resume from system sleep, pciehp checks the
Presence Detect State bit in the Slot Status register to bring up an
occupied slot or bring down an unoccupied slot. Both code paths are
identical, so deduplicate them per Mika's request.
On probe, an additional check is performed to disable power of an
unoccupied slot. This can e.g. happen if power was enabled by BIOS.
It cannot happen once pciehp has taken control, hence is not necessary
on resume: The Slot Control register is set to the same value that it
had on suspend by pci_restore_state(), so if the slot was occupied,
power is enabled and if it wasn't, power is disabled. Should occupancy
have changed during the system sleep transition, power is adjusted by
bringing up or down the slot per the paragraph above.
To allow for deduplication of the presence check, move the power check
to pcie_init(). This seems safer anyway, because right now it is
performed while interrupts are already enabled, and although I can't
think of a scenario where pciehp_power_off_slot() and the IRQ thread
collide, it does feel brittle.
However this means that pcie_init() may now write to the Slot Control
register before the IRQ is requested. If both the CCIE and HPIE bits
happen to be set, pcie_wait_cmd() will wait for an interrupt (instead
of polling the Command Completed bit) and eventually emit a timeout
message. Additionally, if a level-triggered INTx interrupt is used,
the user may see a spurious interrupt splat. Avoid by disabling
interrupts before disabling power. (Normally the HPIE and CCIE bits
should be clear on probe, but conceivably they may already have been
set e.g. by BIOS.)
Signed-off-by: Lukas Wunner <lukas@wunner.de>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Reviewed-by: Mika Westerberg <mika.westerberg@linux.intel.com>
Per Mika's request, add an explicit break to the last case of switch
statements everywhere in pciehp to be more defensive towards future
amendments.
Per Gustavo's request, mark all non-empty implicit fallthroughs with a
comment to silence warnings triggered by -Wimplicit-fallthrough=2.
Signed-off-by: Lukas Wunner <lukas@wunner.de>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Reviewed-by: Mika Westerberg <mika.westerberg@linux.intel.com>
Acked-by: Gustavo A. R. Silva <gustavo@embeddedor.com>
When a PCI device is detected, pdev->is_added is set to 1 and proc and
sysfs entries are created.
When the device is removed, pdev->is_added is checked for one and then
device is detached with clearing of proc and sys entries and at end,
pdev->is_added is set to 0.
is_added and is_busmaster are bit fields in pci_dev structure sharing same
memory location.
A strange issue was observed with multiple removal and rescan of a PCIe
NVMe device using sysfs commands where is_added flag was observed as zero
instead of one while removing device and proc,sys entries are not cleared.
This causes issue in later device addition with warning message
"proc_dir_entry" already registered.
Debugging revealed a race condition between the PCI core setting the
is_added bit in pci_bus_add_device() and the NVMe driver reset work-queue
setting the is_busmaster bit in pci_set_master(). As these fields are not
handled atomically, that clears the is_added bit.
Move the is_added bit to a separate private flag variable and use atomic
functions to set and retrieve the device addition state. This avoids the
race because is_added no longer shares a memory location with is_busmaster.
Link: https://bugzilla.kernel.org/show_bug.cgi?id=200283
Signed-off-by: Hari Vyas <hari.vyas@broadcom.com>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Reviewed-by: Lukas Wunner <lukas@wunner.de>
Acked-by: Michael Ellerman <mpe@ellerman.id.au>
Thunderbolt controllers can be runtime suspended to D3cold to save ~1.5W.
This requires that runtime D3 is allowed on its PCIe ports, so whitelist
them.
The 2015 BIOS cutoff that we've instituted for runtime D3 on PCIe ports
is unnecessary on Thunderbolt because we know that even the oldest
controller, Light Ridge (2010), is able to suspend its ports to D3 just
fine -- specifically including its hotplug ports. And the power saving
should be afforded to machines even if their BIOS predates 2015.
Signed-off-by: Lukas Wunner <lukas@wunner.de>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Reviewed-by: Mika Westerberg <mika.westerberg@linux.intel.com>
Cc: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Cc: Andreas Noever <andreas.noever@gmail.com>
Previously we blacklisted PCIe hotplug ports for runtime D3 because:
(a) Ports handled by the firmware must not be transitioned to D3 by the
OS behind the firmware's back:
https://bugzilla.kernel.org/show_bug.cgi?id=53811
(b) Ports handled natively by the OS lacked runtime D3 support in the
pciehp driver.
We've just rectified the latter, so allow users to manually enable and
test it by passing pcie_port_pm=force on the command line. Vendors are
thus put in a position to validate hotplug ports for runtime D3 and
perhaps we can someday enable it by default, but with a BIOS cutoff date.
Ashok Raj tested runtime D3 on hotplug ports of a SkyLake Xeon-SP in
2017 and encountered Hardware Error NMIs, so this feature clearly cannot
be enabled for everyone yet:
https://lkml.kernel.org/r/20170503180426.GA4058@otc-nc-03
While at it, remove an erroneous code comment I added with 97a90aee5d
("PCI: Consolidate conditions to allow runtime PM on PCIe ports") which
claims that parents of a hotplug port must stay awake lest interrupts
cannot be delivered. That has turned out to be wrong at least for
Thunderbolt hotplug ports.
Signed-off-by: Lukas Wunner <lukas@wunner.de>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Cc: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Cc: Mika Westerberg <mika.westerberg@linux.intel.com>
Cc: Ashok Raj <ashok.raj@intel.com>
Cc: Keith Busch <keith.busch@intel.com>
Cc: Yinghai Lu <yinghai@kernel.org>
When performing a function reset via sysfs, the device's config space is
accessed in places such as pcie_flr() and its MMIO space is accessed e.g.
in reset_ivb_igd(), so ensure accessibility by resuming the device to D0.
Signed-off-by: Lukas Wunner <lukas@wunner.de>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Cc: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Cc: Mika Westerberg <mika.westerberg@linux.intel.com>
Cc: Ashok Raj <ashok.raj@intel.com>
Cc: Keith Busch <keith.busch@intel.com>
Cc: Yinghai Lu <yinghai@kernel.org>
Ensure accessibility of a hotplug port's config space when accessed via
sysfs by resuming its parent to D0.
Signed-off-by: Lukas Wunner <lukas@wunner.de>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Cc: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Cc: Mika Westerberg <mika.westerberg@linux.intel.com>
Cc: Ashok Raj <ashok.raj@intel.com>
Cc: Keith Busch <keith.busch@intel.com>
Cc: Yinghai Lu <yinghai@kernel.org>
pciehp's IRQ thread ensures accessibility of the port by runtime resuming
its parent to D0. However when the slot is enabled/disabled, the port
itself needs to be in D0 because its secondary bus is accessed in:
pciehp_check_link_status(),
pciehp_configure_device() (both called from board_added())
and
pciehp_unconfigure_device() (called from remove_board()).
Thus, acquire a runtime PM ref on enable/disablement of the slot.
Yinghai Lu additionally discovered that some SkyLake servers feature a
Power Controller for their PCIe hotplug ports (PCIe r3.1, sec 6.7.1.8)
which requires the port to be in D0 when invoking
pciehp_power_on_slot() (likewise called from board_added()).
If slot power is turned on while in D3hot, link training later fails:
https://lkml.kernel.org/r/20170205073454.GA253@wunner.de
The spec is silent about such a requirement, but it seems prudent to
assume that any hotplug port with a Power Controller may need this.
The present commit holds a runtime PM ref whenever slot power is turned
on and off, but it doesn't keep the port in D0 as long as slot power is
on. If vendors determine that's necessary, they need to amend pciehp to
acquire a runtime PM ref in pciehp_power_on_slot() and release one in
pciehp_power_off_slot().
Signed-off-by: Lukas Wunner <lukas@wunner.de>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Cc: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Cc: Mika Westerberg <mika.westerberg@linux.intel.com>
Cc: Ashok Raj <ashok.raj@intel.com>
Cc: Keith Busch <keith.busch@intel.com>
Cc: Yinghai Lu <yinghai@kernel.org>
If a hotplug port is able to send an interrupt, one would naively assume
that it is accessible at that moment. After all, if it wouldn't be
accessible, i.e. if its parent is in D3hot and the link to the hotplug
port is thus down, how should an interrupt come through?
It turns out that assumption is wrong at least for Thunderbolt: Even
though its parents are in D3hot, a Thunderbolt hotplug port is able to
signal interrupts. Because the port's config space is inaccessible and
resuming the parents may sleep, the hard IRQ handler has to defer
runtime resuming the parents and reading the Slot Status register to the
IRQ thread.
If the hotplug port uses a level-triggered INTx interrupt, it needs to
be masked until the IRQ thread has cleared the signaled events. For
simplicity, this commit also masks edge-triggered MSI/MSI-X interrupts.
Note that if the interrupt is shared (which can only happen for INTx),
other devices are starved from receiving interrupts until the IRQ thread
is scheduled, has runtime resumed the hotplug port's parents and has
read and cleared the Slot Status register.
That delay is dominated by the 10 ms D3hot->D0 transition time of each
parent port. The worst case is a Thunderbolt downstream port at the
end of a daisy chain: There may be up to six Thunderbolt controllers
in-between it and the root port, each comprising an upstream and
downstream port, plus its own upstream port. That's 13 x 10 = 130 ms.
Possible mitigations are polling the interrupt while it's disabled or
reducing the d3_delay of Thunderbolt ports if possible.
Open code masking of the interrupt instead of requesting it with the
IRQF_ONESHOT flag to minimize the period during which it is masked.
(IRQF_ONESHOT unmasks the IRQ only after the IRQ thread has finished.)
PCIe r4.0 sec 6.7.3.4 states that "If wake generation is required by the
associated form factor specification, a hotplug capable Downstream Port
must support generation of a wakeup event (using the PME mechanism) on
hotplug events that occur when the system is in a sleep state or the
Port is in device state D1, D2, or D3Hot."
This would seem to imply that PME needs to be enabled on the hotplug
port when it is runtime suspended. pci_enable_wake() currently doesn't
enable PME on bridges, it may be necessary to add an exemption for
hotplug bridges there. On "Light Ridge" Thunderbolt controllers, the
PME_Status bit is not set when an interrupt occurs while the hotplug
port is in D3hot, even if PME is enabled. (I've tested this on a Mac
and we hardcode the OSC_PCI_EXPRESS_PME_CONTROL bit to 0 on Macs in
negotiate_os_control(), modifying it to 1 didn't change the behavior.)
(Side note: Section 6.7.3.4 also states that "PME and Hot-Plug Event
interrupts (when both are implemented) always share the same MSI or
MSI-X vector". That would only seem to apply to Root Ports, however
the section never mentions Root Ports, only Downstream Ports. This is
explained in the definition of "Downstream Port" in the "Terms and
Acronyms" section of the PCIe Base Spec: "The Ports on a Switch that
are not the Upstream Port are Downstream Ports. All Ports on a Root
Complex are Downstream Ports.")
Signed-off-by: Lukas Wunner <lukas@wunner.de>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Cc: Mika Westerberg <mika.westerberg@linux.intel.com>
Cc: Ashok Raj <ashok.raj@intel.com>
Cc: Keith Busch <keith.busch@intel.com>
Cc: Yinghai Lu <yinghai@kernel.org>
Upon resume from system sleep, the Slot Control register is written via:
pci_pm_resume_noirq()
pci_pm_default_resume_early()
pci_restore_state()
pci_restore_pcie_state()
PCIe r4.0, sec 6.7.3.2 says that after "issuing a write transaction that
targets any portion of the Port's Slot Control register, [...] software
must wait for [the] command to complete before issuing the next command".
pciehp currently fails to enforce that rule after the above-mentioned
write. Fix it.
(Moving restoration of the Slot Control register to pciehp doesn't seem
to make sense because the other PCIe hotplug drivers may need it as
well.)
Signed-off-by: Lukas Wunner <lukas@wunner.de>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Thunderbolt hotplug ports that were occupied before system sleep resume
with their downstream link in "off" state. Only after the Thunderbolt
controller has reestablished the PCIe tunnels does the link go up.
As a result, a spurious Presence Detect Changed and/or Data Link Layer
State Changed event occurs.
The events are not immediately acted upon because tunnel reestablishment
happens in the ->resume_noirq phase, when interrupts are still disabled.
Also, notification of events may initially be disabled in the Slot
Control register when coming out of system sleep and is reenabled in the
->resume_noirq phase through:
pci_pm_resume_noirq()
pci_pm_default_resume_early()
pci_restore_state()
pci_restore_pcie_state()
It is not guaranteed that the events are acted upon at all: PCIe r4.0,
sec 6.7.3.4 says that "a port may optionally send an MSI when there are
hot-plug events that occur while interrupt generation is disabled, and
interrupt generation is subsequently enabled." Note the "optionally".
If an MSI is sent, pciehp will gratuitously turn the slot off and back
on once the ->resume_early phase has commenced.
If an MSI is not sent, the extant, unacknowledged events in the Slot
Status register will prevent future notification of presence or link
changes.
Commit 13c65840fe ("PCI: pciehp: Clear Presence Detect and Data Link
Layer Status Changed on resume") fixed the latter by clearing the events
in the ->resume phase. Move this to the ->resume_noirq phase to also
fix the gratuitous disable/enablement of the slot.
The commit further restored the Slot Control register in the ->resume
phase, but that's dispensable because as shown above it's already been
done in the ->resume_noirq phase.
Signed-off-by: Lukas Wunner <lukas@wunner.de>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Cc: Mika Westerberg <mika.westerberg@linux.intel.com>
