The address range check may overflow if the memory region is located at
the top of the 32-bit address space. This can e.g. be seen on TK1 if
using the E1000 gigabit Ethernet driver where start and size are both
0x80000000 leading to the following messages:
Apalis TK1 # tftpboot $loadaddr test_file
Using e1000#0 device
TFTP from server 192.168.10.1; our IP address is 192.168.10.2
Filename 'test_file'.
Load address: 0x80408000
Loading: pci_hose_phys_to_bus: invalid physical address
This patch fixes this by changing the order of the addition vs.
subtraction in the range check just like already done in
__pci_hose_bus_to_phys().
Reported-by: Ivan Mercier <ivan.mercier@nexvision.fr>
Signed-off-by: Marcel Ziswiler <marcel.ziswiler@toradex.com>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Reviewed-by: Stephen Warren <swarren@nvidia.com>
Acked-by: Simon Glass <sjg@chromium.org>
When converting between PCI bus and phys addresses, a two pass search
was introduced with preference to non-PCI_REGION_SYS_MEMORY regions.
See commit 2d43e873a2.
However, since PCI_REGION_MEM is defined as 0, the if statement was
always asserted true: ((flags & PCI_REGION_MEM) == PCI_REGION_MEM)
This patch uses PCI_REGION_TYPE bit to check if the region is
PCI_REGION_MEM: ((flags & PCI_REGION_TYPE) == PCI_REGION_MEM)
Signed-off-by: Cheng Gu <chenggu@marvell.com>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
This needs a separate compatible value from Tegra124 since the new HW
version has bugs that would prevent a driver for previous HW versions
from operating at all.
Signed-off-by: Stephen Warren <swarren@nvidia.com>
Signed-off-by: Tom Warren <twarren@nvidia.com>
The board PCI setup code may control regulators that are required simply
to bring up the PCI controller itself (or PLLs, IOs, ... it uses). Move
the call to this function earlier so that all board-provided resources
are ready early enough for everything to work.
Signed-off-by: Stephen Warren <swarren@nvidia.com>
Signed-off-by: Tom Warren <twarren@nvidia.com>
Tegra210's PCIe controller has a bug that requires the PCA (performance
counter) feature to be enabled. If this isn't done, accesses to device
configuration space will hang the chip for tens of seconds. Implement
the workaround.
Signed-off-by: Stephen Warren <swarren@nvidia.com>
Signed-off-by: Tom Warren <twarren@nvidia.com>
The number of cells used by each entry in the DT ranges property is
determined by the #address-cells/#size-cells properties. Fix the code
to respect this.
Signed-off-by: Stephen Warren <swarren@nvidia.com>
Signed-off-by: Tom Warren <twarren@nvidia.com>
Tegra peripherals can generally access a 32-bit physical address space,
and I believe this applies to PCIe. Clip the PCI region that refers to
DRAM so it fits into 32-bits to avoid issues.
Signed-off-by: Stephen Warren <swarren@nvidia.com>
Signed-off-by: Tom Warren <twarren@nvidia.com>
There are two LS series processors are built on ARMv8 Layersacpe
architecture currently, LS2085A and LS1043A. They are based on
ARMv8 core although use different chassis, so create fsl-layerscape
to refactor the common code for the LS series processors which also
paves the way for adding LS1043A platform.
Signed-off-by: Mingkai Hu <Mingkai.Hu@freescale.com>
Signed-off-by: Hou Zhiqiang <B48286@freescale.com>
Signed-off-by: Gong Qianyu <Qianyu.Gong@freescale.com>
Reviewed-by: York Sun <yorksun@freescale.com>
To support graphics card behind a PCI bridge, the bridge control
register (offset 0x3e) in the configuration space must turn on
VGA address forwarding.
Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Acked-by: Simon Glass <sjg@chromium.org>
Currently pci_last_busno() only checks the last bridge device
under the first UCLASS_PCI device. This is not the case when
there are multiple bridge devices.
Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Acked-by: Simon Glass <sjg@chromium.org>
PCI_COMMAND_IO bit must be set for VGA device as it needs to respond
to legacy VGA IO address.
Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Acked-by: Simon Glass <sjg@chromium.org>
The current code returns 0 even if it failed to find or bind a driver. The
caller then has to check the returned device to see if it is NULL. It is
better to return an error code in this case so that it is clear what
happened.
Adjust the code to return -EPERM, indicating that the device was not bound
because it is not needed for pre-relocation use. Add comments so that the
return value is clear.
Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
One debug() statement is missing a newline. The other has a repeated word.
Fix these.
Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
When the auto-configuration process fails for a device (generally due to
lack of memory) we should return the error correctly so that we don't
continue to try memory allocations which will fail.
Adjust the code to check for errors and abort if something goes wrong.
Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
PCI driver currently hangs on mx6qp.
Toggle the reset bit with the appropriate timings to fix the issue.
Based on the FSL kernel driver implementation.
Signed-off-by: Fabio Estevam <fabio.estevam@freescale.com>
Acked-by: Stefano Babic <sbabic@denx.de>
PCI_HEADER_TYPE register (offset 0x0e) bit 7 is an indicator
for multi-function devices. We should mask it off before using
it as the header type.
Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Acked-by: Simon Glass <sjg@chromium.org>
Introduce device_is_on_pci_bus() which can be utilized by driver
to test if a device is on a PCI bus.
Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Acked-by: Simon Glass <sjg@chromium.org>
milliseconds should be written as 'ms' instead of 'mS'.
Signed-off-by: Fabio Estevam <fabio.estevam@freescale.com>
Tested-by: Marek Vasut <marex@denx.de>
At present, until a PCI bus is probed, it cannot be found by its sequence
number unless it has an alias. This is the same with any device.
However with PCI this is more annoying than usual, since bus 0 is always the
same device.
Add a function that tries a little harder to locate PCI bus 0. This means
that PCI enumeration will happen automatically on the first access.
Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Setup mmu-masters property for the PCIe controllers. This would be
used by the Linux SMMU driver, while setting up stream ID table mappings
for the PCIe devices.
Signed-off-by: Varun Sethi <Varun.Sethi@freescale.com>
Reviewed-by: York Sun <yorksun@freescale.com>
If there is no pci device listed in the device tree,
don't bother scanning the device tree.
Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Acked-by: Simon Glass <sjg@chromium.org>
In pci_uclass_child_post_bind(), bdf is extracted from fdt_pci_addr.
Mask bus number before save it to pplat->devfn.
Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Acked-by: Simon Glass <sjg@chromium.org>
Per Intel FSP specification, we should call FSP notify API to
inform FSP that PCI enumeration has been done so that FSP will
do any necessary initialization as required by the chipset's
BIOS Writer's Guide (BWG).
Unfortunately we have to put this call here as with driver model,
the enumeration is all done on a lazy basis as needed, so until
something is touched on PCI it won't happen.
Note we only call this after U-Boot is relocated and root bus has
finished probing.
Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Acked-by: Simon Glass <sjg@chromium.org>
On some platforms pci devices behind bridge need to be probed (eg:
a pci uart on recent x86 chipset) before relocation. But we won't
bind all devices found during the enumeration. Only devices whose
driver with DM_FLAG_PRE_RELOC set will be bound. Any other generic
devices except bridges won't be bound.
Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Acked-by: Simon Glass <sjg@chromium.org>
Currenlty we only set up video framebuffer when VIDEO_VESA driver is
used. With coreboot, VIDEO_COREBOOT driver is used instead. Since we
already saved VESA mode in the VIDEO_COREBOOT driver, now we can also
set up video framebuffer for coreboot before loading Linux kernel.
Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
When booting as a coreboot payload, the framebuffer details are
passed from coreboot via configuration tables. We save these
information into vesa_mode_info structure for future use.
Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
This adds a PCI driver for the controllers found on Marvell MVEBU SoCs.
Besides the driver, this patch also removes the statically defined
PCI MBUS windows. As they are not needed anymore, since this PCIe
driver now creates the windows dynamically.
Tested on Armada XP db-mv784mp-gp eval board using an Intel E1000
PCIe card in all 3 PCIe slots. And on the Armada 38x db-88f6820-gp
eval board using this Intel E1000 PCIe card in the PCIe 0 slot.
This port was done in cooperation with Anton Schubert.
Signed-off-by: Anton Schubert <anton.schubert@gmx.de>
Signed-off-by: Stefan Roese <sr@denx.de>
Cc: Luka Perkov <luka.perkov@sartura.hr>
Cc: Dirk Eibach <eibach@gdsys.de>
This reverts commit df189d9ba3.
Unfortunately this commit breaks chromebook_link because it adds lots of PCI devices
before relocation and there is not enough pre-reloc malloc() memory.
Rathar then increase this memory, revert for now until we figure this out.
Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
These functions allow iteration through all PCI devices including bridges.
The children of each PCI bus are returned in turn. This can be useful for
configuring, checking or enumerating all the devices.
Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
At present there are no PCI functions which allow access to PCI
configuration using a struct udevice. This is a sad situation for driver
model as it makes use of PCI harder. Add these functions.
Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
fdt_addr_t is changed to phys_addr_t. The format in debug should be updated
to %pa to match the type.
Signed-off-by: York Sun <yorksun@freescale.com>
CC: Simon Glass <sjg@chromium.org>
At present the PCI output displays 'Mem' when it allocates memory for a PCI
device, whether it is prefetchable or not. There is a distinction since the
memory comes from separate pools. Use 'Prf' instead of 'Mem' when allocating
prefetchable memory.
Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
This option is not used by any board but appears to still be useful, at least
for testing. With recent commits it does not build, so fix it.
Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
This file does not need its own way of doing debug(). Clean it up to use the
new way.
Signed-off-by: Simon Glass <sjg@chromium.org>
Acked-by: Pavel Machek <pavel@denx.de>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
We should not set up kernel screen_info when the vesa parameters are
insane, otherwise kernel will panic.
Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Acked-by: Simon Glass <sjg@chromium.org>
On some platforms pci devices behind bridge need to be probed (eg:
a pci uart on recent x86 chipset) before relocation. Remove such
limitation so that dm pci can be used before relocation.
Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Acked-by: Simon Glass <sjg@chromium.org>
The patch will initialize PCIe controller on EP mode
1. Setup bar:
bar0 32bit 4K for specific configuration
bar1 32bit 8K for MSIX
bar2 64bit 4K for descriptor of memory
bar4 64bit 1M for DMA memory test
2. Setup iATU:
iATU inbound 0-3 to map bar transaction to memory address
started at CONFIG_SYS_PCI_EP_MEMORY_BASE
iATU outbound 0 to map 4G memory space
Signed-off-by: Minghuan Lian <Minghuan.Lian@freescale.com>
Reviewed-by: York Sun <yorksun@freescale.com>
PCIe extends device's configuration space to 4k and provides
extended capability. The patch adds function to find them.
The code is ported from Linux PCIe driver.
Signed-off-by: Minghuan Lian <Minghuan.Lian@freescale.com>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Reviewed-by: York Sun <yorksun@freescale.com>
Commit aec241d "dm: pci: Use the correct hose when configuring devices"
was an attempt to fix pci bridge device configuration, but unfortunately
that does not work 100%. In pciauto_config_devices(), the fix tried to
call pciauto_config_device() with a ctlr_hose which is supposed to be
the root controller hose, however when walking through a pci topology
with 2 or more pci bridges this logic simply fails.
The call chain is: pciauto_config_devices()->pciauto_config_device()
->dm_pci_hose_probe_bus(). Here the call to dm_pci_hose_probe_bus()
does not make any sense as the given hose is not the bridge device's
hose, instead it is either the root controller's hose (case#1: if it
is the 2nd pci bridge), or the bridge's parent bridge's hose (case#2:
if it is the 3rd pci bridge). In both cases the logic is wrong.
For example, for failing case#1 if the bridge device to config has the
same devfn as one of the devices under the root controller, the call
to pci_bus_find_devfn() will return the udevice of that pci device
under the root controller as the bus, but this is wrong as the udevice
is not a bus which does not contain all the necessary bits associated
with the udevice which causes further failures.
To correctly support pci bridge device configuration, we should still
call pciauto_config_device() with the pci bridge's hose directly.
In order to access valid pci region information, we need to refer to
the root controller simply by a call to pci_bus_to_hose(0) and get the
region information there in the pciauto_prescan_setup_bridge(),
pciauto_postscan_setup_bridge() and pciauto_config_device().
Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
In dm_pci_hose_probe_bus(), pci_bus_find_devfn() is called with a bdf
which includes a bus number, but it really should not as this routine
only expects a device/function encoding.
Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Currently pci_bus_read_config() and pci_bus_write_config() are
called with bus number masked off in the parameter bdf, and bus
number is supposed to be added back in the bridge driver's pci
config read/write ops if the device is behind a pci bridge.
However this logic only works for a pci topology where there is
only one bridge off the root controller. If there is addtional
bridge in the system, the logic will create a non-existent bdf
where its bus number gets accumulated across bridges.
To correct this, we change all pci config read/write routines
to use complete bdf all the way up to the root controller.
Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
In driver model, each pci bridge device has its own hose structure.
hose->first_busno points to the bridge device's device number, so
we should not substract hose->first_busno before programming the
bridge device's primary/secondary/subordinate bus number registers.
Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
It is useful to be able to find the full PCI address (bus, device and
function) for a PCI device. Add a function to provide this.
Adjust the existing code to use this.
Signed-off-by: Simon Glass <sjg@chromium.org>
At present all PCI devices must be present in the device tree in order to
be used. Many or most PCI devices don't require any configuration other than
that which is done automatically by U-Boot. It is inefficent to add a node
with nothing but a compatible string in order to get a device working.
Add a mechanism whereby PCI drivers can be declared along with the device
parameters they support (vendor/device/class). When no suitable driver is
found in the device tree the list of such devices is consulted to determine
the correct driver. If this also fails, then a generic driver is used as
before.
The mechanism used is very similar to that provided by Linux and the header
file defintions are copied from Linux 4.1.
Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Joe Hershberger <joe.hershberger@ni.com>
This patch adds the infrastructure to update device
tree nodes to convey SMMU stream IDs in the device
tree. Fixups are implemented for PCI controllers
initially.
Signed-off-by: Stuart Yoder <stuart.yoder@freescale.com>
Signed-off-by: Prabhakar Kushwaha <prabhakar@freescale.com>
Reviewed-by: York Sun <yorksun@freescale.com>
We should not leave the expansion ROM address window open when there
is not a valid ROM.
Suggested-by: Matt Porter <mporter@konsulko.com>
Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Tested-by: Simon Glass <sjg@chromium.org>
Acked-by: Simon Glass <sjg@chromium.org>
Currently PCI expansion ROM address is assigned by a call to
pciauto_setup_rom() outside of the pci auto config process.
This does not work when expansion ROM is on a device behind
PCI bridge where bridge's memory limit register was already
programmed to a value that does not cover the newly assigned
expansion ROM address. To fix this, we should configure the
ROM address during the auto config process.
Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
Tested-by: Simon Glass <sjg@chromium.org>
Acked-by: Simon Glass <sjg@chromium.org>
