PCI_EXP_CAP is an iProc-specific value, so rename it to IPROC_PCI_EXP_CAP
to make it obvious that it's not related to the generic values like
PCI_EXP_RTCTL, etc. No functional change intended.
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
During soft reset (e.g., "reboot" from Linux) on some iProc-based SOCs, the
LCPLL clock and PERST both go off simultaneously. This seems in accordance
with the PCIe Card Electromechanical spec, r2.0, sec 2.2.3, which says the
clock goes inactive after PERST# goes active, but doesn't specify how long
the clock should be valid after PERST#.
However, we have observed that with the iProc Stingray, some Intel NVMe
endpoints, e.g., the P3700 400GB series, are not detected correctly upon
the next boot sequence unless the clock remains valid for some time after
PERST# is asserted.
Delay 500ms after asserting PERST# before performing a reboot. The 500ms
is experimentally determined.
Signed-off-by: Oza Pawandeep <oza.oza@broadcom.com>
[bhelgaas: changelog, add spec reference, fold in iproc_pcie_shutdown()
export from Arnd Bergmann <arnd@arndb.de>]
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Reviewed-by: Ray Jui <ray.jui@broadcom.com>
Reviewed-by: Scott Branden <scott.branden@broadcom.com>
Configuration Request Retry Status ("CRS") completions are a required part
of PCIe. A PCIe device may respond to config a request with a CRS
completion to indicate that it needs more time to initialize. A Root Port
that receives a CRS completion may automatically retry the request, or it
may treat the request as a failed transaction. For a failed read, it will
likely synthesize all 1's data, i.e., 0xffffffff, to complete the read to
the CPU.
CRS Software Visibility ("CRS SV") is an optional feature. Per PCIe r3.1,
sec 2.3.2, if supported and enabled, a Root Port that receives a CRS
completion for a config read of the Vendor ID will synthesize 0x0001 data
(an invalid Vendor ID) instead of retrying or failing the transaction. The
0x0001 data makes the CRS completion visible to software, so it can perform
other tasks while waiting for the device.
The iProc "Stingray" PCIe controller does not support CRS completions
correctly. From the Stingray PCIe Controller spec:
4.7.3.3. Retry Status On Configuration Cycle
Endpoints are allowed to generate retry status on configuration cycles.
In this case, the RC needs to re-issue the request. The IP does not
handle this because the number of configuration cycles needed will
probably be less than the total number of non-posted operations needed.
When a retry status is received on the User RX interface for a
configuration request that was sent on the User TX interface, it will be
indicated with a completion with the CMPL_STATUS field set to 2=CRS, and
the user will have to find the address and data values and send a new
transaction on the User TX interface. When the internal configuration
space returns a retry status during a configuration cycle (user_cscfg =
1) on the Command/Status interface, the pcie_cscrs will assert with the
pcie_csack signal to indicate the CRS status.
When the CRS Software Visibility Enable register in the Root Control
register is enabled, the IP will return the data value to 0x0001 for the
Vendor ID value and 0xffff (all 1’s) for the rest of the data in the
request for reads of offset 0 that return with CRS status. This is true
for both the User RX Interface and for the Command/Status interface.
When CRS Software Visibility is enabled, the CMPL_STATUS field of the
completion on the User RX Interface will not be 2=CRS and the pcie_cscrs
signal will not assert on the Command/Status interface.
The Stingray hardware never reissues configuration requests when it
receives CRS completions. Contrary to what sec 4.7.3.3 above says, when it
receives a CRS completion, it synthesizes 0xffff0001 data regardless of the
address of the read or the value of the CRS SV enable bit.
This is broken in two ways:
1) When CRS SV is disabled, the Root Port should never synthesize the
0x0001 value. If it receives a CRS completion, it should fail the
transaction and synthesize all 1's data.
2) When CRS SV is enabled, the Root Port should only synthesize 0x0001
data if it receives a CRS completion for a read of the Vendor ID. If it
receives a CRS completion for any other read, it should fail the
transaction and synthesize all 1's data.
This breaks pci_flr_wait(), which reads the Command register and expects to
see all 1's data if the read fails because of CRS completions. On
Stingray, it sees the incorrect 0xffff0001 data instead.
It also breaks config registers that contain the 0xffff0001 value. If we
read such a register, software can't distinguish a CRS completion from the
actual value read from the device.
On Stingray, if we read 0xffff0001 data, assume this indicates a CRS
completion and retry the read for 500ms. If we time out, return all 1's
(0xffffffff) data. Note that this corrupts registers that happen to
contain 0xffff0001.
Stingray advertises CRS SV support in its Root Capabilities register, and
the CRS SV enable bit is writable (even though the hardware ignores it).
Mask out PCI_EXP_RTCAP_CRSVIS so software doesn't try to use CRS SV.
Signed-off-by: Oza Pawandeep <oza.oza@broadcom.com>
[bhelgaas: changelog, add probe-time warning about corruption, don't
advertise CRS SV support, remove duplicate pci_generic_config_read32(),
fix alignment based on patch from Arnd Bergmann <arnd@arndb.de>]
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Factor out the address calculation for memory-mapped config accesses as a
separate function. No functional change intended.
Signed-off-by: Oza Pawandeep <oza.oza@broadcom.com>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Since, through struct pci_host_bridge.map/swizzle_irq hooks, IRQs are now
allocated in the pci_assign_irq() callback automatically, PCI host bridge
drivers can stop relying on pci_fixup_irqs() for IRQ allocation.
Drop pci_fixup_irqs() usage from PCI iproc host bridge driver.
Signed-off-by: Lorenzo Pieralisi <lorenzo.pieralisi@arm.com>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Cc: Scott Branden <sbranden@broadcom.com>
Cc: Ray Jui <rjui@broadcom.com>
Cc: Jon Mason <jonmason@broadcom.com>
The introduction of pci_scan_root_bus_bridge() provides a PCI core API to
scan a PCI root bus backed by an already initialized struct pci_host_bridge
object, which simplifies the bus scan interface and makes the PCI scan root
bus interface easier to generalize as members are added to the struct
pci_host_bridge.
Convert PCI iproc host code to pci_scan_root_bus_bridge() to improve the
PCI root bus scanning interface.
Signed-off-by: Lorenzo Pieralisi <lorenzo.pieralisi@arm.com>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Cc: Scott Branden <sbranden@broadcom.com>
Cc: Ray Jui <rjui@broadcom.com>
Cc: Jon Mason <jonmason@broadcom.com>
The current iproc driver host bridge controller driver requires struct
pci_bus to be created in order to carry out PCI link checks with standard
PCI config space accessors.
This struct pci_bus dependency is fictitious and burdens the driver with
unneeded constraints (eg to use separate APIs to create and scan the root
bus).
Add PCI raw config space accessors to the iproc driver and remove the
fictitious struct pci_bus dependency.
Signed-off-by: Lorenzo Pieralisi <lorenzo.pieralisi@arm.com>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Cc: Scott Branden <sbranden@broadcom.com>
Cc: Ray Jui <rjui@broadcom.com>
Cc: Jon Mason <jonmason@broadcom.com>
Make sure PCIe MPS settings are valid when we enumerate a new hierarchy.
[bhelgaas: changelog]
Signed-off-by: Jon Mason <jon.mason@broadcom.com>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Acked-by: Ray Jui <ray.jui@broadcom.com>
PAXC and PAXCv2 buses do not support legacy IRQs so there is no reason to
even try and map them. Without a change like this, one cannot create VFs
on Nitro ports since legacy interrupts are checked as part of the PCI
device creation process. Testing on PAXC hardware showed that VFs are
properly created with only the change to not set pcie->map_irq, but just to
be safe the change in iproc_pcie_setup() will ensure that pdev_fixup_irq()
will not panic.
Signed-off-by: Andy Gospodarek <gospo@broadcom.com>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Acked-by: Ray Jui <ray.jui@broadcom.com>
In the code to handle PAXB v2 based MSI steering, the logic aligns the MSI
register address to the size of supported inbound mapping range. This is
incorrect since it rounds "up" the starting address to the next aligned
address, but what we want is the starting address to be rounded "down" to
the aligned address.
This patch fixes the issue and allows MSI writes to be properly steered to
the GIC.
Fixes: 4b073155fbd3 ("PCI: iproc: Add support for the next-gen PAXB controller")
Signed-off-by: Ray Jui <ray.jui@broadcom.com>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Add support for the next generation of the iProc PAXB host controller, used
in Stingray.
Signed-off-by: Oza Oza <oza.oza@broadcom.com>
Signed-off-by: Ray Jui <ray.jui@broadcom.com>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Reviewed-by: Scott Branden <scott.branden@broadcom.com>
Add support for inbound DMA mapping. The range of the inbound mapping is
configured by the optional device tree property 'dma-ranges'.
While inbound mapping is done automatically in the ASIC on most iProc-based
SoCs, newer ASICs (e.g., Stingray) require inbound mapping to be configured
explicitly in software.
[bhelgaas: fold in fixes to avoid 32-bit division in iproc_pcie_ib_write()
and uninitialized return value in iproc_pcie_setup_ib() from Arnd Bergmann
<arnd@arndb.de>]
Signed-off-by: Oza Oza <oza.oza@broadcom.com>
Signed-off-by: Ray Jui <ray.jui@broadcom.com>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Reviewed-by: Scott Branden <scott.branden@broadcom.com>
Improve the iProc PCIe outbound mapping code by making it more generic and
removing redundant device tree properties 'brcm,pcie-ob-window-size' and
'brcm,pcie-ob-oarr-size'. The driver is still backward compatible to
device tree binaries with the two properties specified.
The driver now automatically configures the correct mapping window size and
number of mapping windows based on the value of device tree property
'ranges' and the capability of of the iProc PCIe controller.
Signed-off-by: Oza Oza <oza.oza@broadcom.com>
Signed-off-by: Ray Jui <ray.jui@broadcom.com>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Reviewed-by: Scott Branden <scott.branden@broadcom.com>
Add support for the second generation of the iProc PCIe PAXC host
controller.
Signed-off-by: Ray Jui <ray.jui@broadcom.com>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Reviewed-by: Anup Patel <anup.patel@broadcom.com>
Reviewed-by: Scott Branden <scott.branden@broadcom.com>
During enumeration with multi-function EP devices, access to the
configuration space of a non-existent function results in an unsupported
request being returned as expected. By default the PAXB-based iProc PCIe
controller forwards this as an APB error to the host system and that causes
an exception, which is undesired.
Disable this undesired behaviour and let the kernel PCI stack deal with an
access to the non-existent function, in which case a vendor ID of 0xffff is
returned and handled gracefully.
Reported-by: JD Zheng <jiandong.zheng@broadcom.com>
Signed-off-by: Ray Jui <ray.jui@broadcom.com>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Reviewed-by: JD Zheng <jiandong.zheng@broadcom.com>
Reviewed-by: Oza Oza <oza.oza@broadcom.com>
Reviewed-by: Scott Branden <scott.branden@broadcom.com>
The iProc PCIe driver is currently using type IPROC_PCIE_PAXB for the
following SoCs: NS, NSP, Cygnus, NS2, and Pegasus. In fact, the BCMA-based
NS uses a legacy PAXB controller that is slightly different from the PAXB
controller used in the rest of SoCs, e.g., some registers are missing and
it does not require software configuration of outbound/inbound address
mapping.
Add a new type, IPROC_PCIE_PAXB_BCMA, to allow us to properly support the
BCMA-based NS along with other iProc-based SoCs going forward.
Signed-off-by: Ray Jui <ray.jui@broadcom.com>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Reviewed-by: Scott Branden <scott.branden@broadcom.com>
During initialization, the current iProc PCIe host driver resets PAXC and
the downstream internal endpoint device that PAXC connects to. If the
endpoint device is already loaded with firmware and has started running
from the bootloader stage, this downstream reset causes the endpoint device
to stop working.
Signed-off-by: Ray Jui <raj.jui@broadcom.com>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Reviewed-by: Scott Branden <scott.branden@broadcom.com>
As the number of iProc PCIe core registers starts to grow and differ
between different revisions of the iProc PCIe controllers, the
current way of populating each individual unsupported register with
value 'IPROC_PCIE_REG_INVALID' with a table entry has become a bit
messy and is difficult to scale up in the future.
Improve the current driver by populating the invalid entries with code
instead of through individual table entries. This helps to avoid a
significant number of invalid table entries when support for the next
revision of the iProc controller is added.
Signed-off-by: Ray Jui <ray.jui@broadcom.com>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Reviewed-by: Scott Branden <scott.branden@broadcom.com>
We know where the PCIe capability lives in the host bridge's config space;
in fact, we already hard-coded the offset of the Link Control 2 register.
The hard-coded Link Control 2 offset was 0xdc. Link Control 2 is at offset
0x30 into the PCIe capability, so the capability itself must be at
0xdc - 0x30 = 0xac.
Hard-code the PCIe capability offset, which means we don't have to search
for it and we can use the standard definitions for registers within the
capability.
No functional change intended.
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
The callers never pass a null "pcie" pointer (they check for kzalloc
failure), so we don't need to check here. The bus driver should never call
the probe function with a null ->dev pointer, so we don't need to check
that either. No functional change intended.
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Validate iproc_pcie->base for BCMA devices just like we already do for
platform devices in iproc_pcie_pltfm_probe(). No functional change
intended.
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Use a local "struct device *dev" for brevity and consistency with other
drivers. No functional change intended.
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Request host bridge window resources so they appear in ioport_resource and
iomem_resource and are reflected in /proc/ioports and /proc/iomem.
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Commit 943ebae781 ("PCI: iproc: Add PAXC interface support") only allowed
device 0, which is a regression on BCMA-based platforms.
All systems support only one device, a Root Port at 00:00.0, on the root
bus. PAXC-based systems support only the Root Port (00:00.0) and a single
device (with multiple functions) below it, e.g., 01:00.0, 01:00.1, etc.
Non-PAXC systems support arbitrary devices below the Root Port.
[bhelgaas: changelog, fold in removal of MAX_NUM_PAXC_PF check]
Fixes: 943ebae781 ("PCI: iproc: Add PAXC interface support")
Reported-by: Rafal Milecki <zajec5@gmail.com>
Signed-off-by: Ray Jui <rjui@broadcom.com>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Add PCIe MSI support for both PAXB and PAXC interfaces on all iProc-based
platforms.
The iProc PCIe MSI support deploys an event queue-based implementation.
Each event queue is serviced by a GIC interrupt and can support up to 64
MSI vectors. Host memory is allocated for the event queues, and each event
queue consists of 64 word-sized entries. MSI data is written to the lower
16-bit of each entry, whereas the upper 16-bit of the entry is reserved for
the controller for internal processing.
Each event queue is tracked by a head pointer and tail pointer. Head
pointer indicates the next entry in the event queue to be processed by
the driver and is updated by the driver after processing is done.
The controller uses the tail pointer as the next MSI data insertion
point. The controller ensures MSI data is flushed to host memory before
updating the tail pointer and then triggering the interrupt.
MSI IRQ affinity is supported by evenly distributing the interrupts to each
CPU core. MSI vector is moved from one GIC interrupt to another in order
to steer to the target CPU.
Therefore, the actual number of supported MSI vectors is:
M * 64 / N
where M denotes the number of GIC interrupts (event queues), and N denotes
the number of CPU cores.
This iProc event queue-based MSI support should not be used with newer
platforms with integrated MSI support in the GIC (e.g., giv2m or
gicv3-its).
[bhelgaas: fold in Kconfig fixes from Arnd Bergmann <arnd@arndb.de>]
Signed-off-by: Ray Jui <rjui@broadcom.com>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Reviewed-by: Anup Patel <anup.patel@broadcom.com>
Reviewed-by: Vikram Prakash <vikramp@broadcom.com>
Reviewed-by: Scott Branden <sbranden@broadcom.com>
Reviewed-by: Marc Zyngier <marc.zyngier@arm.com>
Traditionally, all iProc PCIe root complexes use PAXB-based wrapper, with
an integrated on-chip Serdes to support external endpoint devices. On
newer iProc platforms, a PAXC-based wrapper is introduced, for connection
with internally emulated PCIe endpoint devices in the ASIC.
Add support for PAXC-based iProc PCIe root complex in the iProc PCIe core
driver. This change factors out common logic between PAXB and PAXC, and
uses tables to store register offsets that are different between PAXB and
PAXC. This allows the driver to be scaled to support subsequent PAXC
revisions in the future.
Signed-off-by: Ray Jui <rjui@broadcom.com>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Reviewed-by: Scott Branden <sbranden@broadcom.com>
The "%pap" format adds a "0x" prefix, so using "0x%pap" results in output
of "0x0x...". Drop the "0x" prefix in the format string.
[bhelgaas: changelog]
Signed-off-by: Dmitry V. Krivenok <krivenok.dmitry@gmail.com>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Acked-by: Ray Jui <rjui@broadcom.com>
Certain SoCs require the PCIe outbound mapping to be configured in
software. Add support for those chips.
[jonmason: Use %pap format when printing size_t to avoid warnings in 32-bit
build.]
[arnd: Use div64_u64() instead of "%" to avoid __aeabi_uldivmod link error
in 32-bit build.]
Signed-off-by: Ray Jui <rjui@broadcom.com>
Signed-off-by: Jon Mason <jonmason@broadcom.com>
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Improve the link detection logic by explicitly querying the link status
register to ensure link is active.
Also force class to PCI_CLASS_BRIDGE_PCI (0x0604) through the host
configuration space register.
Signed-off-by: Ray Jui <rjui@broadcom.com>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Reviewed-by: Anup Patel <anup.patel@broadcom.com>
Reviewed-by: Scott Branden <sbranden@broadcom.com>
The current reset logic does not always properly reset the device. For
example, in the case when the perst_b signal is already de-asserted in the
bootloader, the current reset logic fails to trigger a proper assert ->
de-assert reset sequence.
Fix the issue by always triggering the proper reset sequence.
Also explicitly select the desired reset source, i.e., perst_b, and reduce
the wait time after the device comes out of reset from 250 ms to 100 ms,
based on recommendation from the ASIC team.
Tested-by: Vladimir Dreizin <vdreizin@broadcom.com>
Tested-by: Darren Edamura <dedamura@broadcom.com>
Signed-off-by: Ray Jui <rjui@broadcom.com>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Reviewed-by: Vladimir Dreizin <vdreizin@broadcom.com>
Reviewed-by: Trac Hoang <trhoang@broadcom.com>
Reviewed-by: Scott Branden <sbranden@broadcom.com>
After 459a07721c ("PCI: Build setup-irq.o for arm64"), we build
setup-irq.o for arm64, so we can use pci_fixup_irqs() on both arm and
arm64.
Remove the "#ifdef CONFIG_ARM" around the call to pci_fixup_irqs().
[bhelgaas: changelog]
Signed-off-by: Ray Jui <rjui@broadcom.com>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Add arm64 support to the iProc PCIe driver.
Note that on arm32, bus->sysdata points to the arm32-specific pci_sys_data
struct, and pci_sys_data.private_data contains the iproc_pcie pointer.
For arm64, there's nothing corresponding to pci_sys_data, so we keep the
iproc_pcie pointer directly in bus->sysdata.
In addition, arm64 does IRQ mapping in pcibios_add_device(), so it doesn't
need pci_fixup_irqs() as arm32 does.
Signed-off-by: Ray Jui <rjui@broadcom.com>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Reviewed-by: Scott Branden <sbranden@broadcom.com>
The functions phy_exit() and phy_power_off() test whether their argument is
NULL and then return immediately. Thus the test around the calls is not
needed.
This issue was detected by using the Coccinelle software.
[bhelgaas: also phy_init() and phy_power_on(), as Ray Jui suggested]
[bhelgaas: also remove tests in iproc_pcie_remove()]
Signed-off-by: Markus Elfring <elfring@users.sourceforge.net>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Reviewed-by: Ray Jui <rjui@broadcom.com>
The struct iproc_pcie.resources member was pointing to a stack variable and
is invalid after the registration function returned.
Remove this pointer and add a parameter to the function.
Tested-by: Ray Jui <rjui@broadcom.com>
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Reviewed-by: Ray Jui <rjui@broadcom.com>
The iProc core PCIe driver defaults to using of_irq_parse_and_map_pci() for
IRQ mapping. Add iproc_pcie.map_irq so bus interfaces that don't use
device tree can override this by supplying their own IRQ mapping function.
[bhelgaas: changelog]
Posting: http://lkml.kernel.org/r/1431465781-10753-1-git-send-email-hauke@hauke-m.de
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Reviewed-by: Ray Jui <rjui@broadcom.com.com>
Add support for the Broadcom iProc PCIe controller.
pcie-iproc.c is the common core driver, and a front-end bus interface needs
to be added to support different bus interfaces.
pcie-iproc-platform.c contains the support for the platform bus interface.
Signed-off-by: Ray Jui <rjui@broadcom.com>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Reviewed-by: Scott Branden <sbranden@broadcom.com>
Acked-by: Arnd Bergmann <arnd@arndb.de>
