| Age | Commit message (Collapse) | Author | Files | Lines |
|---|
|
The interrupt affinity spreading mechanism supports to spread out
affinities for one or more interrupt sets. A interrupt set contains one or
more interrupts. Each set is mapped to a specific functionality of a
device, e.g. general I/O queues and read I/O queus of multiqueue block
devices.
The number of interrupts per set is defined by the driver. It depends on
the total number of available interrupts for the device, which is
determined by the PCI capabilites and the availability of underlying CPU
resources, and the number of queues which the device provides and the
driver wants to instantiate.
The driver passes initial configuration for the interrupt allocation via a
pointer to struct irq_affinity.
Right now the allocation mechanism is complex as it requires to have a loop
in the driver to determine the maximum number of interrupts which are
provided by the PCI capabilities and the underlying CPU resources. This
loop would have to be replicated in every driver which wants to utilize
this mechanism. That's unwanted code duplication and error prone.
In order to move this into generic facilities it is required to have a
mechanism, which allows the recalculation of the interrupt sets and their
size, in the core code. As the core code does not have any knowledge about the
underlying device, a driver specific callback is required in struct
irq_affinity, which can be invoked by the core code. The callback gets the
number of available interupts as an argument, so the driver can calculate the
corresponding number and size of interrupt sets.
At the moment the struct irq_affinity pointer which is handed in from the
driver and passed through to several core functions is marked 'const', but for
the callback to be able to modify the data in the struct it's required to
remove the 'const' qualifier.
Add the optional callback to struct irq_affinity, which allows drivers to
recalculate the number and size of interrupt sets and remove the 'const'
qualifier.
For simple invocations, which do not supply a callback, a default callback
is installed, which just sets nr_sets to 1 and transfers the number of
spreadable vectors to the set_size array at index 0.
This is for now guarded by a check for nr_sets != 0 to keep the NVME driver
working until it is converted to the callback mechanism.
To make sure that the driver configuration is correct under all circumstances
the callback is invoked even when there are no interrupts for queues left,
i.e. the pre/post requirements already exhaust the numner of available
interrupts.
At the PCI layer irq_create_affinity_masks() has to be invoked even for the
case where the legacy interrupt is used. That ensures that the callback is
invoked and the device driver can adjust to that situation.
[ tglx: Fixed the simple case (no sets required). Moved the sanity check
for nr_sets after the invocation of the callback so it catches
broken drivers. Fixed the kernel doc comments for struct
irq_affinity and de-'This patch'-ed the changelog ]
Signed-off-by: Ming Lei <[email protected]>
Signed-off-by: Thomas Gleixner <[email protected]>
Acked-by: Marc Zyngier <[email protected]>
Cc: Christoph Hellwig <[email protected]>
Cc: Bjorn Helgaas <[email protected]>
Cc: Jens Axboe <[email protected]>
Cc: [email protected]
Cc: Sagi Grimberg <[email protected]>
Cc: [email protected]
Cc: [email protected]
Cc: Keith Busch <[email protected]>
Cc: Sumit Saxena <[email protected]>
Cc: Kashyap Desai <[email protected]>
Cc: Shivasharan Srikanteshwara <[email protected]>
Link: https://lkml.kernel.org/r/[email protected]
|
|
Commit 0e157e528604 ("PCI/PME: Implement runtime PM callbacks") tried to
solve an issue where the hierarchy immediately wakes up when it is
transitioned into D3cold. However, it turns out to prevent PME
propagation on some systems that do not support D3cold.
I looked more closely at what might cause the immediate wakeup. It happens
when the ACPI power resource of the root port is turned off. The AML code
associated with the _OFF() method of the ACPI power resource starts a PCIe
L2/L3 Ready transition and waits for it to complete. Right after the L2/L3
Ready transition is started the root port receives a PME from the
downstream port.
The simplest hierarchy where this happens looks like this:
00:1d.0 PCIe Root Port
^
|
v
05:00.0 PCIe switch #1 upstream port
06:01.0 PCIe switch #1 downstream hotplug port
^
|
v
08:00.0 PCIe switch #2 upstream port
It seems that the PCIe link between the two switches, before
PME_Turn_Off/PME_TO_Ack is complete for the whole hierarchy, goes
inactive and triggers PME towards the root port bringing it back to D0.
The L2/L3 Ready sequence is described in PCIe r4.0 spec sections 5.2 and
5.3.3 but unfortunately they do not state what happens if DLLSCE is
enabled during the sequence.
Disabling Data Link Layer State Changed event (DLLSCE) seems to prevent
the issue and still allows the downstream hotplug port to notice when a
device is plugged/unplugged.
Link: https://bugzilla.kernel.org/show_bug.cgi?id=202593
Fixes: 0e157e528604 ("PCI/PME: Implement runtime PM callbacks")
Signed-off-by: Mika Westerberg <[email protected]>
Signed-off-by: Bjorn Helgaas <[email protected]>
Reviewed-by: Rafael J. Wysocki <[email protected]>
CC: [email protected] # v4.20+
|
|
The Class Code for subtractive decode PCI-to-PCI bridge is 060401h; add an
entry to make portdrv support this type of bridge. This allows use of PCIe
services on subtractive decode ports.
Signed-off-by: Honghui Zhang <[email protected]>
[bhelgaas: add braces surrounding entry]
Signed-off-by: Bjorn Helgaas <[email protected]>
|
|
The pci_device_id table was technically correct, but unusually formatted,
which made adding entries error-prone. Change the format so it's obvious
how to add entries.
Signed-off-by: Bjorn Helgaas <[email protected]>
|
|
Now that pci-epf-test uses get_features callback and
dw_plat_pcie_epc_features in Designware plat EP driver already indicates
it doesn't support linkup notification and is MSIX capable, remove setting
epc->features which is not used anymore by the endpoint function driver.
Tested-by: Gustavo Pimentel <[email protected]>
Signed-off-by: Kishon Vijay Abraham I <[email protected]>
Signed-off-by: Lorenzo Pieralisi <[email protected]>
|
|
pci_epf_linkup() is intended to be invoked if the EPC supports linkup
notification. Now that pci-epf-test uses get_features callback, which
indicates Rockchip EP driver doesn't support linkup notification, remove
pci_epf_linkup() from Rockchip EP driver.
Tested-by: Gustavo Pimentel <[email protected]>
Signed-off-by: Kishon Vijay Abraham I <[email protected]>
Signed-off-by: Lorenzo Pieralisi <[email protected]>
Acked-by: Shawn Lin <[email protected]>
|
|
pci_epf_linkup() is intended to be invoked if the EPC supports linkup
notification. Now that pci-epf-test uses the get_features() callback,
which indicates Cadence EP driver doesn't support the linkup notification,
remove pci_epf_linkup() from Cadence EP driver.
Tested-by: Gustavo Pimentel <[email protected]>
Signed-off-by: Kishon Vijay Abraham I <[email protected]>
Signed-off-by: Lorenzo Pieralisi <[email protected]>
|
|
Use pci_epc_get_features() to get EPC features such as linkup
notifier support, MSI/MSIX capable, BAR configuration etc and use it
for configuring pci-epf-test. Since these features are now obtained
directly from EPC driver, remove pci_epf_test_data which was initially
added to have EPC features in endpoint function driver.
Tested-by: Gustavo Pimentel <[email protected]>
Signed-off-by: Kishon Vijay Abraham I <[email protected]>
Signed-off-by: Lorenzo Pieralisi <[email protected]>
|
|
It's useless to allocate memory for next BAR if the current BAR is a
64Bit BAR. Stop allocating memory for the next BAR, if the current
BARs flag indicates this is a 64Bit BAR.
Tested-by: Gustavo Pimentel <[email protected]>
Signed-off-by: Kishon Vijay Abraham I <[email protected]>
Signed-off-by: Lorenzo Pieralisi <[email protected]>
|
|
Now that pci_epf_alloc_space() sets BAR MEM TYPE flags as 64Bit or
32Bit based on size, remove setting it in function driver.
Tested-by: Gustavo Pimentel <[email protected]>
Signed-off-by: Kishon Vijay Abraham I <[email protected]>
Signed-off-by: Lorenzo Pieralisi <[email protected]>
|
|
pci_epf_alloc_space() sets the MEM TYPE flags to indicate a 32-bit
Base Address Register irrespective of the size. Fix it here to indicate
64-bit BAR if the size is > 2GB.
Tested-by: Gustavo Pimentel <[email protected]>
Signed-off-by: Kishon Vijay Abraham I <[email protected]>
Signed-off-by: Lorenzo Pieralisi <[email protected]>
|
|
Add a helper function pci_epc_get_first_free_bar() to get the first
unreserved BAR that can be used for endpoint function.
Tested-by: Gustavo Pimentel <[email protected]>
Signed-off-by: Kishon Vijay Abraham I <[email protected]>
Signed-off-by: Lorenzo Pieralisi <[email protected]>
|
|
This reverts commit 0e157e52860441cb26051f131dd0b5ae3187a07b.
Heiner reported that the commit in question prevents his network adapter
from triggering PME and waking up when network cable is plugged.
The commit tried to prevent root port waking up from D3cold immediately but
looks like disabing root port PME interrupt is not the right way to fix
that issue so revert it now. The patch following proposes an alternative
solution to that issue.
Link: https://bugzilla.kernel.org/show_bug.cgi?id=202103
Fixes: 0e157e528604 ("PCI/PME: Implement runtime PM callbacks")
Reported-by: Heiner Kallweit <[email protected]>
Tested-by: Heiner Kallweit <[email protected]>
Signed-off-by: Mika Westerberg <[email protected]>
Signed-off-by: Bjorn Helgaas <[email protected]>
Acked-by: Rafael J. Wysocki <[email protected]>
CC: [email protected] # v4.20+
|
|
Populate ->get_features() dw_pcie_ep_ops to return the EPC features
supported by Cadence PCIe endpoint controller.
Tested-by: Gustavo Pimentel <[email protected]>
Signed-off-by: Kishon Vijay Abraham I <[email protected]>
Signed-off-by: Lorenzo Pieralisi <[email protected]>
|
|
Populate ->get_features() dw_pcie_ep_ops to return the EPC features
supported by Rockchip PCIe endpoint controller.
Tested-by: Gustavo Pimentel <[email protected]>
Signed-off-by: Kishon Vijay Abraham I <[email protected]>
Signed-off-by: Lorenzo Pieralisi <[email protected]>
|
|
Populate ->get_features() dw_pcie_ep_ops to return the EPC features
supported by DRA7xx PCIe endpoint controller.
Tested-by: Gustavo Pimentel <[email protected]>
Signed-off-by: Kishon Vijay Abraham I <[email protected]>
Signed-off-by: Lorenzo Pieralisi <[email protected]>
|
|
Populate ->get_features() dw_pcie_ep_ops to return the EPC features
supported by Designware PCIe endpoint controller.
Tested-by: Gustavo Pimentel <[email protected]>
Signed-off-by: Kishon Vijay Abraham I <[email protected]>
Signed-off-by: Lorenzo Pieralisi <[email protected]>
|
|
Each platform using Designware PCIe core can support different set of
endpoint features. Add a new callback function ->get_features() in
dw_pcie_ep_ops so that each platform using Designware PCIe core can
advertise its supported features to the endpoint function driver.
Tested-by: Gustavo Pimentel <[email protected]>
Signed-off-by: Kishon Vijay Abraham I <[email protected]>
Signed-off-by: Lorenzo Pieralisi <[email protected]>
|
|
Add a new pci_epc_ops ->get_features() to get the features
supported by the EPC. Since EPC can provide different features to
different functions, the ->get_features() ops takes _func_no_ as
an argument.
Tested-by: Gustavo Pimentel <[email protected]>
Signed-off-by: Kishon Vijay Abraham I <[email protected]>
Signed-off-by: Lorenzo Pieralisi <[email protected]>
|
|
Functions copying from/to IO addresses should use the
memcpy_fromio()/memcpy_toio() API rather than plain memcpy().
Fix the issue detected through the sparse tool.
Fixes: 349e7a85b25f ("PCI: endpoint: functions: Add an EP function to test PCI")
Suggested-by: Kishon Vijay Abraham I <[email protected]>
Signed-off-by: Wen Yang <[email protected]>
[[email protected]: updated log]
Signed-off-by: Lorenzo Pieralisi <[email protected]>
Acked-by: Kishon Vijay Abraham I <[email protected]>
CC: Lorenzo Pieralisi <[email protected]>
CC: Bjorn Helgaas <[email protected]>
CC: Gustavo Pimentel <[email protected]>
CC: Niklas Cassel <[email protected]>
CC: Greg Kroah-Hartman <[email protected]>
CC: Cyrille Pitchen <[email protected]>
CC: [email protected]
CC: [email protected]
|
|
This implements the workound described in the NXP IMX7d erratum e10728.
Initial VCO oscillation may fail under corner conditions such as cold
temperature. It causes PCIe PLL to fail to lock in the initialization
phase, which results in the PCIe link failing to come up.
The workaround is to disable Duty-Cycle Corrector (DCC) calibration
after G_RST.
To do this it is necessary to gain access to the undocumented and
currently unused PCIe PHY register bank. A new device tree node of type
"fsl,imx7d-pcie-phy" is created for the PHY block and the existing PCIe
device uses a phandle named "fsl,imx7d-pcie-phy" to point to it.
Signed-off-by: Trent Piepho <[email protected]>
[[email protected]: updated log string, commit log]
Signed-off-by: Lorenzo Pieralisi <[email protected]>
Reviewed-by: Lucas Stach <[email protected]>
|
|
Add debug error message when MSI-X entry control mask bit is set, to help
debug the reason why a MSI-X interrupt is not being triggered.
Signed-off-by: Gustavo Pimentel <[email protected]>
Signed-off-by: Lorenzo Pieralisi <[email protected]>
Cc: Jingoo Han <[email protected]>
Cc: Bjorn Helgaas <[email protected]>
Cc: Joao Pinto <[email protected]>
|
|
Latency Tolerance Reporting (LTR) allows Endpoints and Switch Upstream
Ports to report their latency requirements to upstream components. If ASPM
L1 PM substates are enabled, the LTR information helps determine when a
Link enters L1.2 [1].
Software must set the maximum latency values in the LTR Capability based on
characteristics of the platform, then set LTR Mechanism Enable in the
Device Control 2 register in the PCIe Capability. The device can then use
LTR to report its latency tolerance.
If the device reports a maximum latency value of zero, that means the
device requires the highest possible performance and the ASPM L1.2 substate
is effectively disabled.
We put devices in D3 for suspend, and we assume their internal state is
lost. On resume, previously we did not restore the LTR Capability, but we
did restore the LTR Mechanism Enable bit, so devices would request the
highest possible performance and ASPM L1.2 wouldn't be used.
[1] PCIe r4.0, sec 5.5.1
Link: https://bugzilla.kernel.org/show_bug.cgi?id=201469
Signed-off-by: Bjorn Helgaas <[email protected]>
|
|
Gigabyte X299 DESIGNARE EX motherboard has one PCIe root port that is
connected to an Alpine Ridge Thunderbolt controller. This port has slot
implemented bit set in the config space but other than that it is not
hotplug capable in the sense we are expecting in Linux (it has
dev->is_hotplug_bridge set to 0):
00:1c.4 PCI bridge: Intel Corporation 200 Series PCH PCI Express Root Port #5
Bus: primary=00, secondary=05, subordinate=46, sec-latency=0
Memory behind bridge: 78000000-8fffffff [size=384M]
Prefetchable memory behind bridge: 00003800f8000000-00003800ffffffff [size=128M]
...
Capabilities: [40] Express (v2) Root Port (Slot+), MSI 00
...
SltCap: AttnBtn- PwrCtrl- MRL- AttnInd- PwrInd- HotPlug- Surprise-
Slot #8, PowerLimit 25.000W; Interlock- NoCompl+
SltCtl: Enable: AttnBtn- PwrFlt- MRL- PresDet- CmdCplt- HPIrq- LinkChg-
Control: AttnInd Unknown, PwrInd Unknown, Power- Interlock-
SltSta: Status: AttnBtn- PowerFlt- MRL- CmdCplt- PresDet- Interlock-
Changed: MRL- PresDet+ LinkState+
This system is using ACPI based hotplug to notify the OS that it needs to
rescan the PCI bus (ACPI hotplug).
If there is nothing connected in any of the Thunderbolt ports the root port
will not have any runtime PM active children and is thus automatically
runtime suspended pretty soon after boot by PCI PM core. Now, when a
device is connected the BIOS SMI handler responsible for enumerating newly
added devices is not able to find anything because the port is in D3.
Prevent this from happening by blacklisting PCI power management of this
particular Gigabyte system.
Link: https://bugzilla.kernel.org/show_bug.cgi?id=202031
Reported-by: Kedar A Dongre <[email protected]>
Signed-off-by: Mika Westerberg <[email protected]>
Signed-off-by: Bjorn Helgaas <[email protected]>
Reviewed-by: Rafael J. Wysocki <[email protected]>
|
|
We need the debugfs fixes in here as well.
Signed-off-by: Greg Kroah-Hartman <[email protected]>
|
|
RussianNeuroMancer reported that the Intel 7265 wifi on a Dell Venue 11 Pro
7140 table stopped working after wakeup from suspend and bisected the
problem to 9ab105deb60f ("PCI/ASPM: Disable ASPM L1.2 Substate if we don't
have LTR"). David Ward reported the same problem on a Dell Latitude 7350.
After af8bb9f89838 ("PCI/ACPI: Request LTR control from platform before
using it"), we don't enable LTR unless the platform has granted LTR control
to us. In addition, we don't notice if the platform had already enabled
LTR itself.
After 9ab105deb60f ("PCI/ASPM: Disable ASPM L1.2 Substate if we don't have
LTR"), we avoid using LTR if we don't think the path to the device has LTR
enabled.
The combination means that if the platform itself enables LTR but declines
to give the OS control over LTR, we unnecessarily avoided using ASPM L1.2.
Link: https://bugzilla.kernel.org/show_bug.cgi?id=201469
Fixes: 9ab105deb60f ("PCI/ASPM: Disable ASPM L1.2 Substate if we don't have LTR")
Fixes: af8bb9f89838 ("PCI/ACPI: Request LTR control from platform before using it")
Reported-by: RussianNeuroMancer <[email protected]>
Reported-by: David Ward <[email protected]>
Signed-off-by: Bjorn Helgaas <[email protected]>
CC: [email protected] # v4.18+
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/helgaas/pci
Pull PCI fix from Bjorn Helgaas:
"Work around Synopsys duplicate Device ID (HAPS USB3, NXP i.MX) that
breaks PCIe on I.MX SoCs (Thinh Nguyen)"
* tag 'pci-v5.0-fixes-4' of git://git.kernel.org/pub/scm/linux/kernel/git/helgaas/pci:
PCI: Work around Synopsys duplicate Device ID (HAPS USB3, NXP i.MX)
|
|
The double underscore types are meant for compatibility in userspace
headers which does not apply here. Therefore, change to use the standard
no-underscore types.
The origin of the double underscore types dates back to before the git era
so I was not able to find a commit to see the original justification.
Signed-off-by: Logan Gunthorpe <[email protected]>
Signed-off-by: Bjorn Helgaas <[email protected]>
|
|
There are at least four different parts with the same Vendor and Device
ID ([16c3:abcd]):
1) Synopsys HAPS USB3 controller
2) Synopsys PCIe Root Port in Freescale/NXP i.MX6Q (reported by Lucas)
3) Synopsys PCIe Root Port in Freescale/NXP i.MX6QP (reported by Lukas)
4) Synopsys PCIe Root Port in Freescale/NXP i.MX7D (reported by Trent)
The HAPS USB3 controller has a Class Code of PCI_CLASS_SERIAL_USB_XHCI,
which means the XHCI driver would normally claim it. Previously,
quirk_synopsys_haps() changed the Class Code of all [16c3:abcd] devices,
including the Root Ports, to PCI_CLASS_SERIAL_USB_DEVICE to prevent the
XHCI driver from claiming them so dwc3-haps can claim them instead.
Changing the Class Code of the Root Ports prevents the PCI core from
handling them as bridges, so devices below them don't work.
Restrict the quirk so it only changes the Class Code for devices that start
with the PCI_CLASS_SERIAL_USB_XHCI Class Code, leaving the Root Ports
alone.
Fixes: 03e6742584af ("PCI: Override Synopsys USB 3.x HAPS device class")
Reported-by: Lukas F. Hartmann <[email protected]>
Reported-by: Trent Piepho <[email protected]>
Reported-by: Lucas Stach <[email protected]>
Signed-off-by: Thinh Nguyen <[email protected]>
[bhelgaas: changelog]
Signed-off-by: Bjorn Helgaas <[email protected]>
|
|
Add code needed to support i.MX8MQ variant.
Signed-off-by: Andrey Smirnov <[email protected]>
Signed-off-by: Lorenzo Pieralisi <[email protected]>
Reviewed-by: Lucas Stach <[email protected]>
Cc: Bjorn Helgaas <[email protected]>
Cc: Fabio Estevam <[email protected]>
Cc: Chris Healy <[email protected]>
Cc: Lucas Stach <[email protected]>
Cc: Leonard Crestez <[email protected]>
Cc: "A.s. Dong" <[email protected]>
Cc: Richard Zhu <[email protected]>
|
|
Both i.MX7D and i.MX8MQ have the same behaviour when it comes to
clearing DIRECT_SPEED_CHANGE bit when no speed change occurs, so to
handle variants correctly add a flag instead of checking the IP block
variant.
Signed-off-by: Andrey Smirnov <[email protected]>
[[email protected]: updated log]
Signed-off-by: Lorenzo Pieralisi <[email protected]>
Reviewed-by: Lucas Stach <[email protected]>
Cc: Bjorn Helgaas <[email protected]>
Cc: Fabio Estevam <[email protected]>
Cc: Chris Healy <[email protected]>
Cc: Lucas Stach <[email protected]>
Cc: Leonard Crestez <[email protected]>
Cc: "A.s. Dong" <[email protected]>
Cc: Richard Zhu <[email protected]>
|
|
PCIe PHY IP block on i.MX7D differs from the one used on i.MX6 family,
so none of the code in the current implementation of
imx6_setup_phy_mpll() or imx6_pcie_reset_phy() is applicable.
Introduce IMX6_PCIE_FLAG_IMX6_PHY and check for it in the aforementioned
functions to make sure they are only executed on appropriate PCIe IP
variants.
Tested-by: Trent Piepho <[email protected]>
Signed-off-by: Andrey Smirnov <[email protected]>
[[email protected]: updated log]
Signed-off-by: Lorenzo Pieralisi <[email protected]>
Reviewed-by: Lucas Stach <[email protected]>
Cc: Bjorn Helgaas <[email protected]>
Cc: Fabio Estevam <[email protected]>
Cc: Chris Healy <[email protected]>
Cc: Lucas Stach <[email protected]>
Cc: Leonard Crestez <[email protected]>
Cc: "A.s. Dong" <[email protected]>
Cc: Richard Zhu <[email protected]>
|
|
Introduce driver data struct. This will simplify handling of device
specific differences.
Signed-off-by: Stefan Agner <[email protected]>
[[email protected] reformatted drvdata, to simplify future diffs]
Signed-off-by: Andrey Smirnov <[email protected]>
Signed-off-by: Lorenzo Pieralisi <[email protected]>
Reviewed-by: Lucas Stach <[email protected]>
Cc: Bjorn Helgaas <[email protected]>
Cc: Fabio Estevam <[email protected]>
Cc: Chris Healy <[email protected]>
Cc: Lucas Stach <[email protected]>
Cc: Leonard Crestez <[email protected]>
Cc: "A.s. Dong" <[email protected]>
Cc: Richard Zhu <[email protected]>
|
|
As per Figure 6-3 in PCIe r4.0, sec 6.2.6, ERR_ messages will be forwarded
from the secondary interface to the primary interface, if the SERR# Enable
bit in the Bridge Control register is set.
It seems clear that an ACPI hotplug parameter method (_HPP or _HPX) that
tells us to "enable SERR in the command register" (ACPI v6.2, sec 6.2.8,
6.2.9.1) refers to PCI_COMMAND_SERR, which enables reporting of errors by
the function itself.
For bridges, we also interpreted that to mean we should enable
PCI_BRIDGE_CTL_SERR, which enables *forwarding* of errors by the bridge.
But we didn't enable PCI_BRIDGE_CTL_SERR anywhere else, which means we
never enabled it for non-ACPI systems or ACPI systems that didn't supply
hotplug parameters.
That means errors reported below bridges were often never forwarded up to a
Root Port where they could be signaled via AER.
Enable PCI_BRIDGE_CTL_SERR for all bridges so we can get better error
reporting for downstream devices.
Signed-off-by: Bharat Kumar Gogada <[email protected]>
[bhelgaas: changelog]
Signed-off-by: Bjorn Helgaas <[email protected]>
|
|
The HXT SD4800 PCI controller does not set the Command Completed bit unless
writes to the Slot Command register change "Control" bits.
Add SD4800 to the quirk.
Signed-off-by: Shunyong Yang <[email protected]>
Signed-off-by: Bjorn Helgaas <[email protected]>
CC: Joey Zheng <[email protected]>
|
|
The design of HXT SD4800 ACS feature is the same as QCOM QDF2xxx. Add an
ACS quirk for the SD4800.
Signed-off-by: Shunyong Yang <[email protected]>
[bhelgaas: split to separate patch]
Signed-off-by: Bjorn Helgaas <[email protected]>
Reviewed-by: Sinan Kaya <[email protected]>
CC: Joey Zheng <[email protected]>
|
|
Replace bit rotation operation (1 << bit) with BIT(bit), which
simplifies code reading.
No functional change is intended.
Signed-off-by: Gustavo Pimentel <[email protected]>
Signed-off-by: Lorenzo Pieralisi <[email protected]>
Cc: Lorenzo Pieralisi <[email protected]>
Cc: Joao Pinto <[email protected]>
Cc: Jingoo Han <[email protected]>
|
|
Improve code readability and simplifies mask/unmask operations by
inverting the applied logic (no functional change is intended).
Replace variable name from irq_status to irq_mask, since its goal is to
keep track of which interrupts are masked or not.
Signed-off-by: Gustavo Pimentel <[email protected]>
Signed-off-by: Lorenzo Pieralisi <[email protected]>
Cc: Lorenzo Pieralisi <[email protected]>
Cc: Joao Pinto <[email protected]>
Cc: Jingoo Han <[email protected]>
|
|
Rename variable from data to d to maintain consistency between driver
functions.
No functional change is intended.
Signed-off-by: Gustavo Pimentel <[email protected]>
Signed-off-by: Lorenzo Pieralisi <[email protected]>
Cc: Lorenzo Pieralisi <[email protected]>
Cc: Joao Pinto <[email protected]>
Cc: Jingoo Han <[email protected]>
|
|
Rename variable from data to d to maintain consistency between driver
functions.
No functional change is intended.
Signed-off-by: Gustavo Pimentel <[email protected]>
Signed-off-by: Lorenzo Pieralisi <[email protected]>
Cc: Lorenzo Pieralisi <[email protected]>
Cc: Joao Pinto <[email protected]>
Cc: Jingoo Han <[email protected]>
|
|
Rename variable from data to d to maintain consistency between driver
functions, such as dw_pci_setup_msi_msg().
No functional change is intended.
Signed-off-by: Gustavo Pimentel <[email protected]>
Signed-off-by: Lorenzo Pieralisi <[email protected]>
Cc: Lorenzo Pieralisi <[email protected]>
Cc: Joao Pinto <[email protected]>
Cc: Jingoo Han <[email protected]>
|
|
Rename variable from data to d to maintain consistency between driver
functions, such as dw_msi_mask_irq() and dw_msi_unmask_irq().
No functional change is intended.
Signed-off-by: Gustavo Pimentel <[email protected]>
Signed-off-by: Lorenzo Pieralisi <[email protected]>
Cc: Lorenzo Pieralisi <[email protected]>
Cc: Joao Pinto <[email protected]>
Cc: Jingoo Han <[email protected]>
|
|
Remove unnecessary header include (signal.h) since it doesn't provide
any needed symbols.
Signed-off-by: Gustavo Pimentel <[email protected]>
Signed-off-by: Lorenzo Pieralisi <[email protected]>
Cc: Lorenzo Pieralisi <[email protected]>
Cc: Joao Pinto <[email protected]>
Cc: Jingoo Han <[email protected]>
|
|
Remove unnecessary header include (of_gpio.h) since it doesn't provide
any needed symbols.
Signed-off-by: Gustavo Pimentel <[email protected]>
Signed-off-by: Lorenzo Pieralisi <[email protected]>
Cc: Lorenzo Pieralisi <[email protected]>
Cc: Joao Pinto <[email protected]>
Cc: Jingoo Han <[email protected]>
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/helgaas/pci
Pull PCI fixes from Bjorn Helgaas:
- Revert armada8k GPIO reset change that broke Macchiatobin booting
(Baruch Siach)
- Use actual size config reads on ARM cns3xxx (Koen Vandeputte)
- Fix ARM cns3xxx config write alignment issue (Koen Vandeputte)
- Fix imx6 PHY device link error checking (Leonard Crestez)
- Fix imx6 probe failure on chips without separate PCI power domain
(Leonard Crestez)
* tag 'pci-v5.0-fixes-3' of git://git.kernel.org/pub/scm/linux/kernel/git/helgaas/pci:
Revert "PCI: armada8k: Add support for gpio controlled reset signal"
ARM: cns3xxx: Use actual size reads for PCIe
ARM: cns3xxx: Fix writing to wrong PCI config registers after alignment
PCI: imx: Fix checking pd_pcie_phy device link addition
PCI: imx: Fix probe failure without power domain
|
|
Revert commit 3d71746c42 ("PCI: armada8k: Add support for gpio controlled
reset signal").
That commit breaks boot on Macchiatobin board when a Mellanox NIC is
present in the PCIe slot.
It turns out that full reset cycle requires first comphy serdes
initialization. Reset signal toggle without comphy initialization makes
access to PCI configuration registers stall indefinitely. U-Boot toggles
the Macchiatobin PCIe reset line already at boot, after initializing the
comphy serdes.
So while commit 3d71746c42 ("PCI: armada8k: Add support for gpio controlled
reset signal") enables PCIe on platforms that U-Boot does not touch the
reset line (like Clearfog GT-8K), it breaks PCIe (and boot) on the
Macchiatobin board.
Revert commit 3d71746c42 ("PCI: armada8k: Add support for gpio controlled
reset signal") entirely to fix the Macchiatobin regression.
Reported-by: Sven Auhagen <[email protected]>
Signed-off-by: Baruch Siach <[email protected]>
Signed-off-by: Lorenzo Pieralisi <[email protected]>
|
|
The check on the device_link_add() return value is wrong;
this leads to erroneous code execution, so fix it.
Fixes: 3f7cceeab895 ("PCI: imx: Add multi-pd support")
Signed-off-by: Leonard Crestez <[email protected]>
[[email protected]: updated commit log]
Signed-off-by: Lorenzo Pieralisi <[email protected]>
|
|
On chips without a separate power domain for PCI (such as 6q/6qp) the
imx6_pcie_attach_pd() function incorrectly returns an error.
Fix by returning 0 if dev_pm_domain_attach_by_name() does not find
anything.
Fixes: 3f7cceeab895 ("PCI: imx: Add multi-pd support")
Reported-by: Lukas F.Hartmann <[email protected]>
Signed-off-by: Leonard Crestez <[email protected]>
[[email protected]: updated commit log]
Signed-off-by: Lorenzo Pieralisi <[email protected]>
|
|
Certain PHYs used with PCIe controller can also be used with other
controllers such as USB or SATA. In order to configure the PHY
to work with PCIe controller, invoke phy_set_mode() API with mode
set to PHY_MODE_PCIE.
Signed-off-by: Kishon Vijay Abraham I <[email protected]>
[[email protected]: updated commit log]
Signed-off-by: Lorenzo Pieralisi <[email protected]>
|
|
dra74x/dra76x and dra72x have separate compatible strings. Add support
for these compatible strings in pci-dra7xx driver to perform syscon
configurations required to get x2 mode working.
Signed-off-by: Kishon Vijay Abraham I <[email protected]>
Signed-off-by: Lorenzo Pieralisi <[email protected]>
|