| Age | Commit message (Collapse) | Author | Files | Lines |
|---|
|
Current check for devfn number in mobiveil_pci_valid_device() is
wrong in that it flags as invalid functions present in PCI device 0
in the root bus while it is perfectly valid to access all functions
in PCI device 0 in the root bus.
Update the check in mobiveil_pci_valid_device() to fix the issue.
Fixes: 9af6bcb11e12 ("PCI: mobiveil: Add Mobiveil PCIe Host Bridge IP driver")
Signed-off-by: Hou Zhiqiang <[email protected]>
Signed-off-by: Lorenzo Pieralisi <[email protected]>
Reviewed-by: Minghuan Lian <[email protected]>
|
|
The reset value of Primary, Secondary and Subordinate bus numbers is
zero which is a broken setup.
Program a sensible default value for Primary/Secondary/Subordinate
bus numbers.
Signed-off-by: Hou Zhiqiang <[email protected]>
Signed-off-by: Lorenzo Pieralisi <[email protected]>
Reviewed-by: Minghuan Lian <[email protected]>
Reviewed-by: Subrahmanya Lingappa <[email protected]>
|
|
The irq_set_chained_handler_and_data() call is not dependent on device
tree firmware so it should be moved out of the DT parsing function for
clarity.
Signed-off-by: Hou Zhiqiang <[email protected]>
[[email protected]: rewritten commit log]
Signed-off-by: Lorenzo Pieralisi <[email protected]>
Reviewed-by: Minghuan Lian <[email protected]>
Reviewed-by: Subrahmanya Lingappa <[email protected]>
|
|
The host initializing sequence does not depend on the PCIe link being
up, so move the link up check right before the code scanning the root
bus for enumeration.
Signed-off-by: Hou Zhiqiang <[email protected]>
Signed-off-by: Lorenzo Pieralisi <[email protected]>
Reviewed-by: Minghuan Lian <[email protected]>
Reviewed-by: Subrahmanya Lingappa <[email protected]>
|
|
Fix up the Class Code field in PCI configuration space and set it to
PCI_CLASS_BRIDGE_PCI.
Move the Class Code fixup to function mobiveil_host_init() where
it belongs.
Fixes: 9af6bcb11e12 ("PCI: mobiveil: Add Mobiveil PCIe Host Bridge IP driver")
Signed-off-by: Hou Zhiqiang <[email protected]>
Signed-off-by: Lorenzo Pieralisi <[email protected]>
Reviewed-by: Minghuan Lian <[email protected]>
Reviewed-by: Subrahmanya Lingappa <[email protected]>
|
|
The inbound and outbound windows have completely separate control
registers sets in the host controller MMIO space. Windows control
register are accessed through an MMIO base address and an offset
that depends on the window index.
Since inbound and outbound windows control registers are completely
separate there is no real need to use different window indexes in the
inbound/outbound windows initialization routines to prevent clashing.
To fix this inconsistency, change the MEM inbound window index to 0,
mirroring the outbound window set-up.
Signed-off-by: Hou Zhiqiang <[email protected]>
[[email protected]: update commit log]
Signed-off-by: Lorenzo Pieralisi <[email protected]>
Reviewed-by: Minghuan Lian <[email protected]>
Reviewed-by: Subrahmanya Lingappa <[email protected]>
|
|
As the .map_bus() use the WIN_NUM_0 for CFG transactions,
it is appropriate to pass WIN_NUM_0 explicitly when initializing
the CFG outbound window rather than implicitly relying on the
ob_wins_configure counter.
Signed-off-by: Hou Zhiqiang <[email protected]>
Signed-off-by: Lorenzo Pieralisi <[email protected]>
Reviewed-by: Minghuan Lian <[email protected]>
Reviewed-by: Subrahmanya Lingappa <[email protected]>
|
|
Code that executes the resource list traversal does not need to
delete any node therefore using the *_safe() API version is
useless.
Replace function resource_list_for_each_entry_safe() with the
resource_list_for_each_entry() counterpart.
Signed-off-by: Hou Zhiqiang <[email protected]>
Signed-off-by: Lorenzo Pieralisi <[email protected]>
Reviewed-by: Minghuan Lian <[email protected]>
Reviewed-by: Subrahmanya Lingappa <[email protected]>
|
|
The outbound memory windows PCI base addresses should be taken
from the 'ranges' property of DT node to setup MEM/IO outbound
windows decoding correctly instead of being hardcoded to zero.
Update the code to retrieve the PCI base address for each range
and use it to program the outbound windows address decoders
Fixes: 9af6bcb11e12 ("PCI: mobiveil: Add Mobiveil PCIe Host Bridge IP driver")
Signed-off-by: Hou Zhiqiang <[email protected]>
Signed-off-by: Lorenzo Pieralisi <[email protected]>
Reviewed-by: Minghuan Lian <[email protected]>
Reviewed-by: Subrahmanya Lingappa <[email protected]>
|
|
The Mobiveil internal MSI controller requires separate target addresses,
one per MSI vector; this is clearly incompatible with the Multiple MSI
feature, which requires the same target address for all vectors
requested by an endpoint (ie the Message Address field in the MSI
Capability structure), so the multi MSI feature is clearly not
supported by the host controller driver.
Remove the flag MSI_FLAG_MULTI_PCI_MSI and with it multi MSI support,
fixing the misconfiguration.
Fixes: 1e913e58335f ("PCI: mobiveil: Add MSI support")
Signed-off-by: Hou Zhiqiang <[email protected]>
[[email protected]: commit log]
Signed-off-by: Lorenzo Pieralisi <[email protected]>
Reviewed-by: Minghuan Lian <[email protected]>
|
|
It is confusing to have two sets of functions to read/write
registers, some with csr_readl()/csr_writel(), while others with
read_paged_register()/write_paged_register().
In the register space the lower 3KB of 4KB PCIe configure space can be
accessed directly and higher 1KB through a simple paging mechanism.
Unify the register accessors in csr_readl() and csr_writel() by
comparing the register offset with page access boundary 3KB in the
accessor internal so that the paging mechanism is hidden behind
the csr_read()/write() common function calls.
Signed-off-by: Hou Zhiqiang <[email protected]>
Signed-off-by: Lorenzo Pieralisi <[email protected]>
Reviewed-by: Minghuan Lian <[email protected]>
Reviewed-by: Subrahmanya Lingappa <[email protected]>
|
|
* pm-sleep:
PM: sleep: Drop dev_pm_skip_next_resume_phases()
ACPI: PM: Drop unused function and function header
ACPI: PM: Introduce "poweroff" callbacks for ACPI PM domain and LPSS
ACPI: PM: Simplify and fix PM domain hibernation callbacks
PCI: PM: Simplify bus-level hibernation callbacks
PM: ACPI/PCI: Resume all devices during hibernation
kernel: power: swap: use kzalloc() instead of kmalloc() followed by memset()
PM: sleep: Update struct wakeup_source documentation
drivers: base: power: remove wakeup_sources_stats_dentry variable
PM: suspend: Rename pm_suspend_via_s2idle()
PM: sleep: Show how long dpm_suspend_start() and dpm_suspend_end() take
PM: hibernate: powerpc: Expose pfn_is_nosave() prototype
|
|
The imx6 driver can be used on imx6sx without enabling support for
imx6q or imx7d but the "depends on" condition doesn't allow that.
Instead of making the condition even longer just make it depend on
"ARCH_MXC || COMPILE_TEST" instead.
Signed-off-by: Leonard Crestez <[email protected]>
Signed-off-by: Lorenzo Pieralisi <[email protected]>
Acked-by: Lucas Stach <[email protected]>
|
|
Fix a use-after-free in hv_eject_device_work().
Fixes: 05f151a73ec2 ("PCI: hv: Fix a memory leak in hv_eject_device_work()")
Signed-off-by: Dexuan Cui <[email protected]>
Signed-off-by: Lorenzo Pieralisi <[email protected]>
Reviewed-by: Michael Kelley <[email protected]>
Cc: [email protected]
|
|
The PCI Tegra controller conversion to a device tree configurable
driver in commit d1523b52bff3 ("PCI: tegra: Move PCIe driver
to drivers/pci/host") implied that code for the driver can be
compiled in for a kernel supporting multiple platforms.
Unfortunately, a blind move of the code did not check that some of the
quirks that were applied in arch/arm (eg enabling Relaxed Ordering on
all PCI devices - since the quirk hook erroneously matches PCI_ANY_ID
for both Vendor-ID and Device-ID) are now applied in all kernels that
compile the PCI Tegra controlled driver, DT and ACPI alike.
This is completely wrong, in that enablement of Relaxed Ordering is only
required by default in Tegra20 platforms as described in the Tegra20
Technical Reference Manual (available at
https://developer.nvidia.com/embedded/downloads#?search=tegra%202 in
Section 34.1, where it is mentioned that Relaxed Ordering bit needs to
be enabled in its root ports to avoid deadlock in hardware) and in the
Tegra30 platforms for the same reasons (unfortunately not documented
in the TRM).
There is no other strict requirement on PCI devices Relaxed Ordering
enablement on any other Tegra platforms or PCI host bridge driver.
Fix this quite upsetting situation by limiting the vendor and device IDs
to which the Relaxed Ordering quirk applies to the root ports in
question, reported above.
Signed-off-by: Vidya Sagar <[email protected]>
[[email protected]: completely rewrote the commit log/fixes tag]
Signed-off-by: Lorenzo Pieralisi <[email protected]>
Acked-by: Thierry Reding <[email protected]>
|
|
Driver checks for link up three times before giving up, each retry
attempt is printed as an error. Letting users know that PCIe link is
down and in the process of being brought up again is for debug, not an
error condition.
Signed-off-by: Manikanta Maddireddy <[email protected]>
Signed-off-by: Lorenzo Pieralisi <[email protected]>
Acked-by: Thierry Reding <[email protected]>
|
|
Tegra PCIe has fixed per port SFIO line to signal PERST#, which can be
controlled by AFI port register. However, if a platform routes a
different GPIO to the PCIe slot, then port register cannot control it.
Add support for GPIO based PERST# signal for such platforms. GPIO number
comes from per port PCIe device tree node. PCIe driver probe doesn't
fail if per port "reset-gpios" property is not populated, so platforms
that require this workaround must make sure that the DT property is not
missed in the corresponding device tree.
Link: https://lore.kernel.org/linux-pci/[email protected]/
Signed-off-by: Manikanta Maddireddy <[email protected]>
[[email protected]: squashed in fix in Link]
Signed-off-by: Lorenzo Pieralisi <[email protected]>
Acked-by: Thierry Reding <[email protected]>
|
|
The SR-IOV specification requires both PFs and VFs to implement a PCIe
capability. Generally this is sufficient to assume extended config space
is present, but we generally also perform additional tests to make sure the
extended config space is reachable and not simply an alias of standard
config space. For a VF to exist extended config space must be accessible
on the PF, therefore we can also assume it to be accessible on the VF.
This enables a micro performance optimization previously implemented in
commit 975bb8b4dc93 ("PCI/IOV: Use VF0 cached config space size for other
VFs") to speed up probing of VFs.
Signed-off-by: Alex Williamson <[email protected]>
Signed-off-by: Bjorn Helgaas <[email protected]>
Reviewed-by: Kuppuswamy Sathyanarayanan <[email protected]>
Cc: KarimAllah Ahmed <[email protected]>
Cc: Hao Zheng <[email protected]>
|
|
Revert 975bb8b4dc93 ("PCI/IOV: Use VF0 cached config space size for other
VFs"), which attempted to cache the config space size from the first VF to
re-use for subsequent VFs.
The cached value was determined prior to discovering the PCIe capability on
the VF, which resulted in the first VF reporting the correct config space
size (4K), as it has a special case through pci_cfg_space_size(), while all
the other VFs only reported 256 bytes. As this was only a performance
optimization, we're better off without it.
Fixes: 975bb8b4dc93 ("PCI/IOV: Use VF0 cached config space size for other VFs")
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Alex Williamson <[email protected]>
Signed-off-by: Bjorn Helgaas <[email protected]>
Cc: KarimAllah Ahmed <[email protected]>
Cc: Kuppuswamy Sathyanarayanan <[email protected]>
Cc: Hao Zheng <[email protected]>
|
|
The driver name in /proc/bus/pci/devices can be printed without a printf
format specification, so use seq_puts() instead of seq_printf().
This issue was detected by using the Coccinelle software.
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Markus Elfring <[email protected]>
[bhelgaas: commit log]
Signed-off-by: Bjorn Helgaas <[email protected]>
|
|
Drivers that use dma_virt_ops were meant to be rejected when testing
compatibility for P2PDMA.
This check got inadvertently dropped in one of the later versions of the
original patchset, so add it back.
Fixes: 52916982af48 ("PCI/P2PDMA: Support peer-to-peer memory")
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Logan Gunthorpe <[email protected]>
Signed-off-by: Bjorn Helgaas <[email protected]>
|
|
After a previous change causing all runtime-suspended PCI devices
to be resumed before creating a snapshot image of memory during
hibernation, it is not necessary to worry about the case in which
them might be left in runtime-suspend any more, so get rid of the
code related to that from bus-level PCI hibernation callbacks.
Signed-off-by: Rafael J. Wysocki <[email protected]>
Reviewed-by: Mika Westerberg <[email protected]>
Reviewed-by: Hans de Goede <[email protected]>
|
|
Both the PCI bus type and the ACPI PM domain avoid resuming
runtime-suspended devices with DPM_FLAG_SMART_SUSPEND set during
hibernation (before creating the snapshot image of system memory),
but that turns out to be a mistake. It leads to functional issues
and adds complexity that's hard to justify.
For this reason, resume all runtime-suspended PCI devices and all
devices in the ACPI PM domains before creating a snapshot image of
system memory during hibernation.
Fixes: 05087360fd7a (ACPI / PM: Take SMART_SUSPEND driver flag into account)
Fixes: c4b65157aeef (PCI / PM: Take SMART_SUSPEND driver flag into account)
Link: https://lore.kernel.org/linux-acpi/[email protected]/T/#maf065fe6e4974f2a9d79f332ab99dfaba635f64c
Reported-by: Robert R. Howell <[email protected]>
Tested-by: Robert R. Howell <[email protected]>
Signed-off-by: Rafael J. Wysocki <[email protected]>
Reviewed-by: Mika Westerberg <[email protected]>
Reviewed-by: Hans de Goede <[email protected]>
|
|
If "hotplug_bridges == 0", "!dev->is_hotplug_bridge" is always true, so the
loop that divides the remaining resources among hotplug-capable bridges
does nothing.
Check for "hotplug_bridges == 0" earlier, so we don't even have to compute
the amount of remaining resources. No functional change intended.
Link: https://lore.kernel.org/r/PS2P216MB0642C7A485649D2D787A1C6F80000@PS2P216MB0642.KORP216.PROD.OUTLOOK.COM
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Nicholas Johnson <[email protected]>
Signed-off-by: Bjorn Helgaas <[email protected]>
Reviewed-by: Logan Gunthorpe <[email protected]>
Reviewed-by: Mika Westerberg <[email protected]>
|
|
Reorder pci_bus_distribute_available_resources() to group related code
together. No functional change intended.
Link: https://lore.kernel.org/r/PS2P216MB0642C7A485649D2D787A1C6F80000@PS2P216MB0642.KORP216.PROD.OUTLOOK.COM
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Nicholas Johnson <[email protected]>
Signed-off-by: Bjorn Helgaas <[email protected]>
Reviewed-by: Mika Westerberg <[email protected]>
Reviewed-by: Logan Gunthorpe <[email protected]>
|
|
The functionality is identical to the one currently open coded in
p2pdma.c.
Signed-off-by: Christoph Hellwig <[email protected]>
Reviewed-by: Ira Weiny <[email protected]>
Reviewed-by: Logan Gunthorpe <[email protected]>
Tested-by: Logan Gunthorpe <[email protected]>
Reviewed-by: Dan Williams <[email protected]>
Signed-off-by: Jason Gunthorpe <[email protected]>
|
|
Passing the actual typed structure leads to more understandable code
vs just passing the ref member.
Reported-by: Logan Gunthorpe <[email protected]>
Signed-off-by: Christoph Hellwig <[email protected]>
Reviewed-by: Logan Gunthorpe <[email protected]>
Reviewed-by: Jason Gunthorpe <[email protected]>
Reviewed-by: Dan Williams <[email protected]>
Tested-by: Dan Williams <[email protected]>
Signed-off-by: Jason Gunthorpe <[email protected]>
|
|
The dev_pagemap is a growing too many callbacks. Move them into a
separate ops structure so that they are not duplicated for multiple
instances, and an attacker can't easily overwrite them.
Signed-off-by: Christoph Hellwig <[email protected]>
Reviewed-by: Logan Gunthorpe <[email protected]>
Reviewed-by: Jason Gunthorpe <[email protected]>
Reviewed-by: Dan Williams <[email protected]>
Tested-by: Dan Williams <[email protected]>
Signed-off-by: Jason Gunthorpe <[email protected]>
|
|
|
|
The new route handling in ip_mc_finish_output() from 'net' overlapped
with the new support for returning congestion notifications from BPF
programs.
In order to handle this I had to take the dev_loopback_xmit() calls
out of the switch statement.
The aquantia driver conflicts were simple overlapping changes.
Signed-off-by: David S. Miller <[email protected]>
|
|
Export all configuration space access APIs and also other APIs to
support host controller drivers of dwc core based implementations while
adding support for .remove() hook to build their respective drivers as
modules.
Signed-off-by: Vidya Sagar <[email protected]>
Signed-off-by: Lorenzo Pieralisi <[email protected]>
Acked-by: Gustavo Pimentel <[email protected]>
|
|
Cleanup DBI read and write APIs by removing leading "__" (underscore)
from their names as there is no reason to have leading underscores
in the first place in the function definition.
Remove dbi/dbi2 base address parameters as the same behaviour can be
obtained through read and write APIs. Since dw_pcie_{readl/writel}_dbi()
APIs can't be used for ATU read/write as ATU base address could be
different from DBI base address, implement ATU read/write APIs using ATU
base address without using dw_pcie_{readl/writel}_dbi() APIs.
Signed-off-by: Vidya Sagar <[email protected]>
Signed-off-by: Lorenzo Pieralisi <[email protected]>
Acked-by: Jingoo Han <[email protected]>
|
|
Add an API to group all the tasks to be done to de-initialize host which
can then be called by any dwc core based driver implementations
while adding .remove() support in their respective drivers.
Signed-off-by: Vidya Sagar <[email protected]>
Signed-off-by: Lorenzo Pieralisi <[email protected]>
Acked-by: Gustavo Pimentel <[email protected]>
|
|
In pci_pm_complete() there are checks to decide whether or not to
resume devices that were left in runtime-suspend during the preceding
system-wide transition into a sleep state. They involve checking the
current power state of the device and comparing it with the power
state of it set before the preceding system-wide transition, but the
platform component of the device's power state is not handled
correctly in there.
Namely, on platforms with ACPI, the device power state information
needs to be updated with care, so that the reference counters of
power resources used by the device (if any) are set to ensure that
the refreshed power state of it will be maintained going forward.
To that end, introduce a new ->refresh_state() platform PM callback
for PCI devices, for asking the platform to refresh the device power
state data and ensure that the corresponding power state will be
maintained going forward, make it invoke acpi_device_update_power()
(for devices with ACPI PM) on platforms with ACPI and make
pci_pm_complete() use it, through a new pci_refresh_power_state()
wrapper function.
Fixes: a0d2a959d3da (PCI: Avoid unnecessary resume after direct-complete)
Signed-off-by: Rafael J. Wysocki <[email protected]>
Reviewed-by: Mika Westerberg <[email protected]>
|
|
If otherwise unrelated PCI devices share ACPI power resources turning
them on causes the devices to enter D0uninitialized power state which may
cause problems.
For example in Intel Ice Lake two root ports (RP0 and RP1), Thunderbolt
controller (NHI) and xHCI controller all share power resources as can be
ween in the topology below where power resources are marked with []:
Host bridge
|
+- RP0 ---\
+- RP1 ---|--+--> [TBT]
+- NHI --/ |
| |
| v
+- xHCI --> [D3C]
In a situation where all devices sharing the power resources are in
D3cold (the power resources are turned off) and for example the
Thunderbolt controller is runtime resumed resulting that the power
resources are turned on. This means that the other devices sharing them
(RP0, RP1 and xHCI) are transitioned into D0uninitialized state. If they
were configured to trigger wake (PME) on a certain event that
configuration gets lost after reset so we would need to re-initialize
them to get the wakeup working as expected again. To do so we would need
to runtime resume all of them to make sure their registers get restored
properly before we can runtime suspend them again.
Since we just added concept of "_PR0 dependent device" we can solve this
by calling the relevant add/remove functions when the PCI device is bind
to its ACPI representation. If it has power resources the PCI device
will be added as dependent device to them and runtime resumed whenever
they are physically turned on. This should make sure PCI core can
reconfigure wakes after the device is transitioned into D0uninitialized.
Signed-off-by: Mika Westerberg <[email protected]>
Signed-off-by: Rafael J. Wysocki <[email protected]>
|
|
The ACPI power state returned by acpi_device_get_power() may depend on
the configuration of ACPI power resources in the system which may change
any time after acpi_device_get_power() has returned, unless the
reference counters of the ACPI power resources in question are set to
prevent that from happening. Thus it is invalid to use acpi_device_get_power()
in acpi_pci_get_power_state() the way it is done now and the value of
the ->power.state field in the corresponding struct acpi_device objects
(which reflects the ACPI power resources reference counting, among other
things) should be used instead.
As an example where this becomes an issue is Intel Ice Lake where the
Thunderbolt controller (NHI), two PCIe root ports (RP0 and RP1) and xHCI
all share the same power resources. The following picture with power
resources marked with [] shows the topology:
Host bridge
|
+- RP0 ---\
+- RP1 ---|--+--> [TBT]
+- NHI --/ |
| |
| v
+- xHCI --> [D3C]
Here TBT and D3C are the shared ACPI power resources. ACPI _PR3() method
of the devices in question returns either TBT or D3C or both.
Say we runtime suspend first the root ports RP0 and RP1, then NHI. Now
since the TBT power resource is still on when the root ports are runtime
suspended their dev->current_state is set to D3hot. When NHI is runtime
suspended TBT is finally turned off but state of the root ports remain
to be D3hot. Now when the xHCI is runtime suspended D3C gets also turned
off. PCI core thus has power states of these devices cached in their
dev->current_state as follows:
RP0 -> D3hot
RP1 -> D3hot
NHI -> D3cold
xHCI -> D3cold
If the user now runs lspci for instance, the result is all 1's like in
the below output (00:07.0 is the first root port, RP0):
00:07.0 PCI bridge: Intel Corporation Device 8a1d (rev ff) (prog-if ff)
!!! Unknown header type 7f
Kernel driver in use: pcieport
In short the hardware state is not in sync with the software state
anymore. The exact same thing happens with the PME polling thread which
ends up bringing the root ports back into D0 after they are runtime
suspended.
For this reason, modify acpi_pci_get_power_state() so that it uses the
ACPI device power state that was cached by the ACPI core. This makes the
PCI device power state match the ACPI device power state regardless of
state of the shared power resources which may still be on at this point.
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Mika Westerberg <[email protected]>
Signed-off-by: Rafael J. Wysocki <[email protected]>
|
|
There are platforms that do not call pm_set_suspend_via_firmware(),
so pm_suspend_via_firmware() returns 'false' on them, but the power
states of PCI devices (PCIe ports in particular) are changed as a
result of powering down core platform components during system-wide
suspend. Thus the pm_suspend_via_firmware() checks in
pci_pm_suspend_noirq() and pci_pm_resume_noirq() introduced by
commit 3e26c5feed2a ("PCI: PM: Skip devices in D0 for suspend-to-
idle") are not sufficient to determine that devices left in D0
during suspend will remain in D0 during resume and so the bus-level
power management can be skipped for them.
For this reason, introduce a new global suspend flag,
PM_SUSPEND_FLAG_NO_PLATFORM, set it for suspend-to-idle only
and replace the pm_suspend_via_firmware() checks mentioned above
with checks against this flag.
Fixes: 3e26c5feed2a ("PCI: PM: Skip devices in D0 for suspend-to-idle")
Reported-by: Jon Hunter <[email protected]>
Tested-by: Jon Hunter <[email protected]>
Signed-off-by: Rafael J. Wysocki <[email protected]>
Tested-by: Mika Westerberg <[email protected]>
Reviewed-by: Mika Westerberg <[email protected]>
|
|
According to the PCI Local Bus specification Revision 3.0,
section 6.8.1.3 (Message Control for MSI), endpoints that
are Multiple Message Capable as defined by bits [3:1] in
the Message Control for MSI can request a number of vectors
that is power of two aligned.
As specified in section 6.8.1.6 "Message data for MSI", the Multiple
Message Enable field (bits [6:4] of the Message Control register)
defines the number of low order message data bits the function is
permitted to modify to generate its system software allocated
vectors.
The MSI controller in the Xilinx NWL PCIe controller supports a number
of MSI vectors specified through a bitmap and the hwirq number for an
MSI, that is the value written in the MSI data TLP is determined by
the bitmap allocation.
For instance, in a situation where two endpoints sitting on
the PCI bus request the following MSI configuration, with
the current PCI Xilinx bitmap allocation code (that does not
align MSI vector allocation on a power of two boundary):
Endpoint #1: Requesting 1 MSI vector - allocated bitmap bits 0
Endpoint #2: Requesting 2 MSI vectors - allocated bitmap bits [1,2]
The bitmap value(s) corresponds to the hwirq number that is programmed
into the Message Data for MSI field in the endpoint MSI capability
and is detected by the root complex to fire the corresponding
MSI irqs. The value written in Message Data for MSI field corresponds
to the first bit allocated in the bitmap for Multi MSI vectors.
The current Xilinx NWL MSI allocation code allows a bitmap allocation
that is not a power of two boundaries, so endpoint #2, is allowed to
toggle Message Data bit[0] to differentiate between its two vectors
(meaning that the MSI data will be respectively 0x0 and 0x1 for the two
vectors allocated to endpoint #2).
This clearly aliases with the Endpoint #1 vector allocation, resulting
in a broken Multi MSI implementation.
Update the code to allocate MSI bitmap ranges with a power of two
alignment, fixing the bug.
Fixes: ab597d35ef11 ("PCI: xilinx-nwl: Add support for Xilinx NWL PCIe Host Controller")
Suggested-by: Marc Zyngier <[email protected]>
Signed-off-by: Bharat Kumar Gogada <[email protected]>
[[email protected]: updated commit log]
Signed-off-by: Lorenzo Pieralisi <[email protected]>
Acked-by: Marc Zyngier <[email protected]>
|
|
|
|
There is an arbitrary difference between the prototypes of
bus_find_device() and class_find_device() preventing their callers
from passing the same pair of data and match() arguments to both of
them, which is the const qualifier used in the prototype of
class_find_device(). If that qualifier is also used in the
bus_find_device() prototype, it will be possible to pass the same
match() callback function to both bus_find_device() and
class_find_device(), which will allow some optimizations to be made in
order to avoid code duplication going forward. Also with that, constify
the "data" parameter as it is passed as a const to the match function.
For this reason, change the prototype of bus_find_device() to match
the prototype of class_find_device() and adjust its callers to use the
const qualifier in accordance with the new prototype of it.
Cc: Alexander Shishkin <[email protected]>
Cc: Andrew Lunn <[email protected]>
Cc: Andreas Noever <[email protected]>
Cc: Arnd Bergmann <[email protected]>
Cc: Bjorn Helgaas <[email protected]>
Cc: Corey Minyard <[email protected]>
Cc: Christian Borntraeger <[email protected]>
Cc: David Kershner <[email protected]>
Cc: "David S. Miller" <[email protected]>
Cc: David Airlie <[email protected]>
Cc: Felipe Balbi <[email protected]>
Cc: Frank Rowand <[email protected]>
Cc: Grygorii Strashko <[email protected]>
Cc: Harald Freudenberger <[email protected]>
Cc: Hartmut Knaack <[email protected]>
Cc: Heiko Stuebner <[email protected]>
Cc: Jason Gunthorpe <[email protected]>
Cc: Jonathan Cameron <[email protected]>
Cc: "James E.J. Bottomley" <[email protected]>
Cc: Len Brown <[email protected]>
Cc: Mark Brown <[email protected]>
Cc: Michael Ellerman <[email protected]>
Cc: Michael Jamet <[email protected]>
Cc: "Martin K. Petersen" <[email protected]>
Cc: Peter Oberparleiter <[email protected]>
Cc: Sebastian Ott <[email protected]>
Cc: Srinivas Kandagatla <[email protected]>
Cc: Yehezkel Bernat <[email protected]>
Cc: [email protected]
Acked-by: Corey Minyard <[email protected]>
Acked-by: David Kershner <[email protected]>
Acked-by: Mark Brown <[email protected]>
Acked-by: Rafael J. Wysocki <[email protected]>
Acked-by: Srinivas Kandagatla <[email protected]>
Acked-by: Wolfram Sang <[email protected]> # for the I2C parts
Acked-by: Rob Herring <[email protected]>
Signed-off-by: Suzuki K Poulose <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/helgaas/pci
Pull PCI fix from Bjorn Helgaas:
"If an IOMMU is present, ignore the P2PDMA whitelist we added for v5.2
because we don't yet know how to support P2PDMA in that case (Logan
Gunthorpe)"
* tag 'pci-v5.2-fixes-1' of git://git.kernel.org/pub/scm/linux/kernel/git/helgaas/pci:
PCI/P2PDMA: Ignore root complex whitelist when an IOMMU is present
|
|
Minor SPDX change conflict.
Signed-off-by: David S. Miller <[email protected]>
|
|
Drivers may rely on pci_disable_link_state() having disabled certain
ASPM link states. If OS can't control ASPM then pci_disable_link_state()
turns into a no-op w/o informing the caller. The driver therefore may
falsely assume the respective ASPM link states are disabled.
Let pci_disable_link_state() propagate errors to the caller, enabling
the caller to react accordingly.
Signed-off-by: Heiner Kallweit <[email protected]>
Acked-by: Bjorn Helgaas <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
|
|
Prevent auto-enabling of bridges reallocation when the FW tells us that the
initial configuration must be preserved for a given host bridge.
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Benjamin Herrenschmidt <[email protected]>
Signed-off-by: Bjorn Helgaas <[email protected]>
|
|
With CONFIG_PROVE_LOCKING=y, using sysfs to remove a bridge with a device
below it causes a lockdep warning, e.g.,
# echo 1 > /sys/class/pci_bus/0000:00/device/0000:00:00.0/remove
============================================
WARNING: possible recursive locking detected
...
pci_bus 0000:01: busn_res: [bus 01] is released
The remove recursively removes the subtree below the bridge. Each call
uses a different lock so there's no deadlock, but the locks were all
created with the same lockdep key so the lockdep checker can't tell them
apart.
Mark the "remove" sysfs attribute with __ATTR_IGNORE_LOCKDEP() as it is
safe to ignore the lockdep check between different "remove" kernfs
instances.
There's discussion about a similar issue in USB at [1], which resulted in
356c05d58af0 ("sysfs: get rid of some lockdep false positives") and
e9b526fe7048 ("i2c: suppress lockdep warning on delete_device"), which do
basically the same thing for USB "remove" and i2c "delete_device" files.
[1] https://lore.kernel.org/r/[email protected]
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Marek Vasut <[email protected]>
[bhelgaas: trim commit log, details at above links]
Signed-off-by: Bjorn Helgaas <[email protected]>
Cc: Geert Uytterhoeven <[email protected]>
Cc: Phil Edworthy <[email protected]>
Cc: Simon Horman <[email protected]>
Cc: Tejun Heo <[email protected]>
Cc: Wolfram Sang <[email protected]>
|
|
In Tegra210 AFI design has clamp value for the BIAS pad as 0, which keeps
the bias pad in non power down mode. This is leading to power consumption
of 2 mW in BIAS pad, even if the PCIe partition is powergated. To avoid
unnecessary power consumption, put PEX CLK & BIAS pads in deep power down
mode when PCIe partition is power gated.
Signed-off-by: Manikanta Maddireddy <[email protected]>
Signed-off-by: Lorenzo Pieralisi <[email protected]>
Acked-by: Thierry Reding <[email protected]>
|
|
Tegra186 and Tegra30 have three PCIe root ports. AFI_PEX2_CTRL register
is defined for third root port. Offset of this register in Tegra186 is
different from Tegra30, so add the offset as part of SoC data structure.
Signed-off-by: Manikanta Maddireddy <[email protected]>
Signed-off-by: Lorenzo Pieralisi <[email protected]>
Acked-by: Thierry Reding <[email protected]>
|
|
PRSNT_MAP bit field is programmed to update the slot present status.
PRSNT_SENSE IRQ is triggered when this bit field is programmed, which is
not an error. Add a new if condition to trap PRSNT_SENSE code and print
it with debug log level.
Signed-off-by: Manikanta Maddireddy <[email protected]>
Signed-off-by: Lorenzo Pieralisi <[email protected]>
Acked-by: Thierry Reding <[email protected]>
|
|
Cacheable upstream transactions are supported in Tegra20 and Tegra186
only.
AFI_CACHE_BAR_{0,1}_{ST,SZ} registers are available in Tegra20 to
support cacheable upstream transactions. In Tegra186, AFI_AXCACHE
register is defined instead of AFI_CACHE_BAR_{0,1}_{ST,SZ} to be in line
with its memory subsystem design.
Therefore, program AFI_CACHE_BAR_{0,1}_{ST,SZ} registers only for Tegra20.
Signed-off-by: Manikanta Maddireddy <[email protected]>
[[email protected]: updated commit log]
Signed-off-by: Lorenzo Pieralisi <[email protected]>
Acked-by: Thierry Reding <[email protected]>
|
|
Disable controllers which failed to bring the link up and configure
CLKREQ# signals of these controllers as GPIO. This is required to avoid
CLKREQ# signal of inactive controllers interfering with PLLE power down
sequence.
PCIE_CLKREQ_GPIO bits are defined only in Tegra186, however programming
these bits in other SoCs doesn't cause any side effects. Program these
bits for all Tegra SoCs to avoid a conditional check.
Signed-off-by: Manikanta Maddireddy <[email protected]>
Signed-off-by: Lorenzo Pieralisi <[email protected]>
Acked-by: Thierry Reding <[email protected]>
|