This fixes a problem, where booting Linux using distro boot will
sometimes lead to an invalid instruction exception on the main hart. The
secondary harts are not affected and boot Linux successfully. The root
cause of this problem is a stack overflow on the main hart.
With distro boot, the current default stack size of 8KiB on RISC-V is
not sufficient and will cause a stack overflow. The stacks are allocated
sequentially. In the case of a stack overflow the stack of the main hart
can reach into that of another hart and be corrupted.
The stack overflow previously did not cause any problems, because only
stack frames, which are not used anymore since the hart enters Linux,
were corrupted. Starting with GCC 9, the stack usage has decreased. Now,
only the most recent stack frame overflows into the stack of a secondary
hart and is corrupted. The illegal instruction exception is caused by
the secondary hart overwriting the return address in the stack frame of
the main hart with an address that does not include valid code.
Increase the default stack size of each hart to 16KiB to avoid this
problem.
Reported-by: Aurelien Jarno <aurelien@aurel32.net>
Signed-off-by: Lukas Auer <lukas.auer@aisec.fraunhofer.de>
Tested-by: David Abdurachmanov <david.abdurachmanov@sifive.com>
Tested-by: Aurelien Jarno <aurelien@aurel32.net>
Reviewed-by: Rick Chen <rick@andestech.com>
The commit 1b42ab3eda ("ARM: DRA7: Fixup DSPEVE, IVA and GPU clock
frequencies based on OPP") added the core logic to update the kernel
device-tree blob to adjust the DSP, IVA and GPU DPLL clocks based on
a one-time OPP choice selected in U-Boot for most of the DRA7xx/AM57xx
family of SoCs.
The DSPs on DRA76xP/DRA77xP SoCs (DRA76x ACD package SoCs) though
provide a higher performance and can run at a higher clock frequency
of 850 MHz at OPP_HIGH instead of 750 MHz. Fix up the logic to use the
correct clock rates on these SoCs. Note that this higher clock rate is
not applicable to other Jacinto 6 Plus SoCs (DRA75xP/DRA74xP SoCs or
AM574x SoCs) that follow the ABZ package.
Signed-off-by: Suman Anna <s-anna@ti.com>
Reviewed-by: Lokesh Vutla <lokeshvutla@ti.com>
On NixOS, cross compiled kernels have long suffixes that cause them to
exceed the current maximum path length. The PXE/TFTP max path length is
used for extlinux.conf support as well, which is where this problem
usually manifest's itself.
Signed-off-by: Ben Wolsieffer <benwolsieffer@gmail.com>
Acked-by: Joe Hershberger <joe.hershberger@ni.com>
Enable configs related to DMA and Ethernet so as to support networking at
U-Boot prompt
Signed-off-by: Vignesh Raghavendra <vigneshr@ti.com>
Acked-by: Joe Hershberger <joe.hershberger@ni.com>
Reviewed-by: Grygorii Strashko <grygorii.strashko@ti.com>
Add DT nodes related to DMA and CPSW to -u-boot.dtsi to get networking
up on J721e EVM.
Signed-off-by: Vignesh Raghavendra <vigneshr@ti.com>
Acked-by: Joe Hershberger <joe.hershberger@ni.com>
Reviewed-by: Grygorii Strashko <grygorii.strashko@ti.com>
Get flow ID information for RX DMA channel using dma_get_cfg() interface
instead of reading from DT. This is required in order to avoid DT update
whenever there is change in the range of flow ID allocated to the host.
Signed-off-by: Vignesh Raghavendra <vigneshr@ti.com>
Acked-by: Joe Hershberger <joe.hershberger@ni.com>
Reviewed-by: Grygorii Strashko <grygorii.strashko@ti.com>
Implement dma_get_cfg() interface to pass flow id information for DMA
clients to use. This is needed because on K3 SoCs, CPSW (ethernet) and
UDMA (DMA provider) support "flows" within a given RX DMA channel. This
allows different network packets to be segregated while using same RX
DMA channel. In order for basic ethernet to work, CPSW slave must be
aware of the flow ID allocated for the RX channel by the DMA driver.
This interface allows CPSW to query flow ID from DMA provider and
configure it in CPSW HW.
Signed-off-by: Vignesh Raghavendra <vigneshr@ti.com>
Acked-by: Joe Hershberger <joe.hershberger@ni.com>
Reviewed-by: Grygorii Strashko <grygorii.strashko@ti.com>
Sometimes, there would be a need to exchange data between DMA provider
and DMA client which are very specific to DMA driver of the SoC/platform
and are not generic enough to be put into struct dma. Therefore, introduce
dma_get_cfg() interface to get DMA provider specific data from client
device. Clients can use unique configuration ID flags to get different
configuration data from DMA driver.
Signed-off-by: Vignesh Raghavendra <vigneshr@ti.com>
Acked-by: Joe Hershberger <joe.hershberger@ni.com>
Reviewed-by: Grygorii Strashko <grygorii.strashko@ti.com>
Fixes: e766849713 ("net: ftgmac100: convert the RX/TX descriptor arrays")
Signed-off-by: Cédric Le Goater <clg@kaod.org>
Reviewed-by: Joel Stanley <joel@jms.id.au>
Acked-by: Joe Hershberger <joe.hershberger@ni.com>
This bus is used to access internal SoC PHYs. These PHYs are configured
by the ENETC driver directly, but it's useful to have command line access
to this MDIO to debug the system especially when using new external PHYs.
Signed-off-by: Alex Marginean <alexandru.marginean@nxp.com>
Acked-by: Joe Hershberger <joe.hershberger@ni.com>
On some boards MDIO may get stuck if it detects echo on the line. This is
a know hardware issue, there is a board fix for it. In case we're running
on a board that doesn't have the fix, we don't want to loop here forever
and freeze U-Boot.
Signed-off-by: Alex Marginean <alexandru.marginean@nxp.com>
Acked-by: Joe Hershberger <joe.hershberger@ni.com>
This reduces the time needed to establish a link as we don't reset the link
each time the interface is used. Our Link capabilities do not change at
run-time so there is no need to re-apply PHY configuration each time.
Signed-off-by: Alex Marginean <alexandru.marginean@nxp.com>
Acked-by: Joe Hershberger <joe.hershberger@ni.com>
The original code enabled link speeds up to 1Gbps, but the interface can
go up to 2.5G, enable that speed to in PHY AN mask.
Signed-off-by: Alex Marginean <alexandru.marginean@nxp.com>
Acked-by: Joe Hershberger <joe.hershberger@ni.com>
Uses the new dm_eth_phy_connect helper to connect to the PHY to simplify
the code.
Signed-off-by: Alex Marginean <alexandru.marginean@nxp.com>
Acked-by: Joe Hershberger <joe.hershberger@ni.com>
The function connects an ethernet device to a PHY using DT information.
This API is only available for eth devices with an associated device tree
node.
Signed-off-by: Alex Marginean <alexandru.marginean@nxp.com>
Acked-by: Joe Hershberger <joe.hershberger@ni.com>
Renamed dm_mdio_phy_connect arguments dev to mdiodev and addr to phyaddr
for a bit more clarity and consistency with the following patches.
Also use NULL instead of 0 on error return path.
Signed-off-by: Alex Marginean <alexandru.marginean@nxp.com>
Acked-by: Joe Hershberger <joe.hershberger@ni.com>
Commit b618b37076 ("net: Convert CONFIG_TFTP_BLOCKSIZE to Kconfig")
accidentally set the default *option* TFTP block size to 512 bytes, even
though the comment in the code says that this is a terrible choice. Most
boards didn't define the symbol before, so they got the default block size
of 1468 bytes before, but now use 512 bytes, which is also the fallback.
This leads to both abysmal performance and a lot of hashes printed
on the screen (one character for every 5K), which is both annoying and
slow over serial links.
Set the default block size in Kconfig back to the value it had before.
This improves TFTP performance from 2.8 MB/s to 6.9 MB/s on a Pine64.
Fixes: b618b37076 ("net: Convert CONFIG_TFTP_BLOCKSIZE to Kconfig")
Signed-off-by: Andre Przywara <andre.przywara@arm.com>
Acked-by: Joe Hershberger <joe.hershberger@ni.com>
A couple of optional properties have been introduced for Aquantia PHY
allowing the driver to set up wiring related configuration points that
are otherwise driven by firmware.
Signed-off-by: Alex Marginean <alexandru.marginean@nxp.com>
Acked-by: Joe Hershberger <joe.hershberger@ni.com>
It defines that PHY nodes must be children on MDIO bus nodes and defines
the only required property in U-Boot, reg. This property along with the
example provided are copied over from Linux.
Signed-off-by: Alex Marginean <alexandru.marginean@nxp.com>
Acked-by: Joe Hershberger <joe.hershberger@ni.com>
In some cases the link on the system interface of the aquantia PHY comes up
after the link on line interface. The link state loop only checks the line
side, which may result in first packet sent being lost.
Use aquantia_link_is_up instead, which checks both system and line side on
gen 2/3 PHYs to avoid losing the 1st packet.
Signed-off-by: Alex Marginean <alexandru.marginean@nxp.com>
Acked-by: Joe Hershberger <joe.hershberger@ni.com>
Aquantia PHYs have a SMBus interface mostly used for debug. The addresses
on this interface are normally set up by PHY firmware, but depending on the
board they may end up not being unique. Add an optional DT property used
to change SMBus address if needed.
Signed-off-by: Alex Marginean <alexandru.marginean@nxp.com>
Acked-by: Joe Hershberger <joe.hershberger@ni.com>
MDI pins up to the RJ45 connector may be reversed on the board and the
default PHY configuration applied by firmware may or may not match that.
Add an optional DT property to configure MDI reversal for this case.
Signed-off-by: Alex Marginean <alexandru.marginean@nxp.com>
Acked-by: Joe Hershberger <joe.hershberger@ni.com>
If PHY is not ready for data by the time _config is called, reconfigure the
PHY system interface to use the proper protocol based on phydev->interface,
just in case the defaults set by PHY firmware don't match current
configuration.
Signed-off-by: Florin Laurentiu Chiculita <florinlaurentiu.chiculita@nxp.com>
Signed-off-by: Alex Marginean <alexandru.marginean@nxp.com>
Acked-by: Joe Hershberger <joe.hershberger@ni.com>
Uses the data field in phy_driver structure to identify the PHY generation.
This is useful for custom configuration as non-generic PHY registers are
not 100% compatible between generations.
Signed-off-by: Alex Marginean <alexandru.marginean@nxp.com>
Acked-by: Joe Hershberger <joe.hershberger@ni.com>
The PHY supports XFI and USXGMII, the notable difference being that USX AN
is enabled for USXGMII. Legacy code uses XGMII for any 10G proto and
detects whether USX AN should be enabled or not using a PHY status
register. Keep that functionality too, so we don't break existing drivers.
Signed-off-by: Razvan Ionut Cirjan <razvanionut.cirjan@nxp.com>
Signed-off-by: Alex Marginean <alexandru.marginean@nxp.com>
Acked-by: Joe Hershberger <joe.hershberger@ni.com>
This is useful to carry custom information between the driver structure
associated with a specific HW and the driver code.
Signed-off-by: Alex Marginean <alexandru.marginean@nxp.com>
Acked-by: Joe Hershberger <joe.hershberger@ni.com>
Drivers currently use XGMII for XFI and USXGMII and, where needed, use
other information to identify the actual protocol on the board. With these
two defined drivers can now rely on DT phy-mode property.
Signed-off-by: Alex Marginean <alexandru.marginean@nxp.com>
Acked-by: Joe Hershberger <joe.hershberger@ni.com>
Fix vendor name in MV88E61xx option description.
Signed-off-by: Anatolij Gustschin <agust@denx.de>
Reviewed-by: Chris Packham <judge.packham@gmail.com>
Tested-by: Chris Packham <judge.packham@gmail.com>
Acked-by: Joe Hershberger <joe.hershberger@ni.com>
Support probing and init for 88E6071 switch.
Signed-off-by: Anatolij Gustschin <agust@denx.de>
Reviewed-by: Chris Packham <judge.packham@gmail.com>
Tested-by: Chris Packham <judge.packham@gmail.com>
Acked-by: Joe Hershberger <joe.hershberger@ni.com>
On mv88E6071 the 'EDet' field offset, width and sense control
bits are different, adjust the driver to init the PHY control
register as needed. This fixes not working link detection and
tftp transfers.
Signed-off-by: Anatolij Gustschin <agust@denx.de>
Reviewed-by: Chris Packham <judge.packham@gmail.com>
Tested-by: Chris Packham <judge.packham@gmail.com>
Acked-by: Joe Hershberger <joe.hershberger@ni.com>
On 88E6071 chip the port status register bit field offsets
for duplex and link bits differ. Extend the driver to use
88E6071 specific offset values. The width of bit fields for
speed status differ, too. Adapt for proper port speed
detection on 88E6071.
Signed-off-by: Anatolij Gustschin <agust@denx.de>
Reviewed-by: Chris Packham <judge.packham@gmail.com>
Tested-by: Chris Packham <judge.packham@gmail.com>
Acked-by: Joe Hershberger <joe.hershberger@ni.com>
Extend the driver to init switch register offsets from variables
instead of compile time macros and enable detection of 88E6071 and
compatible devices. Ethernet transfer (e.g. tftp) does not work yet,
so enable the registration of the 'indirect mii' bus for easier PHY
register access by 'mii' command.
Signed-off-by: Anatolij Gustschin <agust@denx.de>
Acked-by: Joe Hershberger <joe.hershberger@ni.com>
The KSZ8081 PHY has a factory test mode which is set at the de-assertion
of the reset line based on the RXER (KSZ8081RNA/RND) or TXC
(KSZ8081MNX/RNB) pin. If a pull-down is missing, or if the pin has a
pull-up, the factory test mode should be cleared by manually writing a 0
(according to the datasheet).
Create another ksz8081_config function to handle this case.
Suggested-by: Antoine Tenart <antoine.tenart@bootlin.com>
Signed-off-by: Nicolas Ferre <nicolas.ferre@microchip.com>
Acked-by: Joe Hershberger <joe.hershberger@ni.com>
device_probe() may fail in which case the seq_id will be -1. Don't
display these devices during startup. While this is only a cosmetic
change, the return value of eth_initialize() will also change to the
actual number of available devices. The return value is only used in
spl_net to decide whether there are any devices to boot from. So
returning only available devices is also more correct in that case.
Signed-off-by: Michael Walle <michael@walle.cc>
Acked-by: Joe Hershberger <joe.hershberger@ni.com>
Aquantia quad-phys may take longer to initialize. This commit adds
a polling mechanism for a global alarm bit that tells if phy init
sequence is completed.
Signed-off-by: Florin Chiculita <florinlaurentiu.chiculita@nxp.com>
Acked-by: Joe Hershberger <joe.hershberger@ni.com>
This allows passing arbitrary addresses through macb_miiphy_read and
macb_miiphy_write, therefore enabling the mii command to access
all mdio bus devices instead of only the defined phy.
Signed-off-by: Josef Holzmayr <holzmayr@rsi-elektrotechnik.de>
Acked-by: Joe Hershberger <joe.hershberger@ni.com>
To support accessing arbitrary addresses the mii/mdio bus it is
necessary that the macb_mdio_read and macb_mdio_write functions
do not implicitly use the address of the connected phy.
The function signature is extended according to the Linux kernel
equivalent.
Signed-off-by: Josef Holzmayr <holzmayr@rsi-elektrotechnik.de>
Acked-by: Joe Hershberger <joe.hershberger@ni.com>
I've noticed that in most cases when genphy_update_link() is called, the
ethernet driver (mt7628-eth in this case) fails with the first ethernet
packets. Resulting in a timeout of the first tftp command. Increasing
the delay in the link check look from 1 to 50 ms and moving it below the
BMSR register read fixes this issue, resulting in a stable ethernet
traffic, even after initial link autonogotiation.
Signed-off-by: Stefan Roese <sr@denx.de>
Cc: Weijie Gao <weijie.gao@mediatek.com>
Cc: Joe Hershberger <joe.hershberger@ni.com>
Acked-by: Joe Hershberger <joe.hershberger@ni.com>
* Add Kconfig help describing the purpose of each command.
* Add CONFIG_CMD_MDIO so it could be selected individually, as
it doesn't depend on the mii command.
* Add Kconfig imply to mii to automatically select the mdio
command.
Signed-off-by: Ramon Fried <rfried.dev@gmail.com>
Acked-by: Joe Hershberger <joe.hershberger@ni.com>
This version of the RTL-8168 chip can be found on some add-in cards sold
by CSL-Computer GmbH & Co. KG. The chip isn't special in any way, but it
needs to have the ChipCmd register programmed after the DMA descriptors
have been set up, so make sure that happens by adding an entry to the
chip information table.
Signed-off-by: Thierry Reding <treding@nvidia.com>
Acked-by: Joe Hershberger <joe.hershberger@ni.com>
Changes to mvpp2.c require the MVMDIO module which in turn uses
DM_MDIO.
Signed-off-by: Nevo Hed <nhed+github@starry.com>
Reviewed-by: Ramon Fried <rfried.dev@gmail.com>
Acked-by: Joe Hershberger <joe.hershberger@ni.com>
This commit ports mvpp2 to use the recently introduced Marvell MDIO
(MVMDIO) driver. It removes direct interaction with the SMI & XSMI
busses. This commit is based in part on earlier work by
Ken Ma <make@marvell.com> in Marvell's own downstream repo:
https://github.com/MarvellEmbeddedProcessors/u-boot-marvell/commit/c81dc39.
The above refrenced work was based on an MVMDIO implementation that
never made it into U-Boot. With this patch the mvpp2 driver switches
to use the new MVMDIO driver that is based on a more universal
mdio-uclass implementation.
Signed-off-by: Nevo Hed <nhed+github@starry.com>
Reviewed-by: Ramon Fried <rfried.dev@gmail.com>
Acked-by: Joe Hershberger <joe.hershberger@ni.com>
Based on upstream-linux
See https://github.com/torvalds/linux/commit/f66b2aff.
However made the XSMI register window 0x16 (22) bytes per my reading
of the functional spec. Similar commits in Marvels own repo bump it
to 0x200 (512) bytes but I did not see the reasoning for that.
https://github.com/MarvellEmbeddedProcessors/u-boot-marvell/commit/4d932b4.
Also added device-name attributes to prevent ambiguity in the `mdio`
command.
Signed-off-by: Nevo Hed <nhed+github@starry.com>
Reviewed-by: Ramon Fried <rfried.dev@gmail.com>
Acked-by: Joe Hershberger <joe.hershberger@ni.com>
If the phy doesn't match with any existing u-boot drivers, the phy
framework will connect it to the generic one which uid ==
0xffffffff. In this case, act as if the phy wouldn't be declared in
dts. Otherwise, in case of 3310 (for which the driver doesn't exist)
the link is marked as always down. Removing phy entry from dts in case
of 3310 is not a good option because it is required for the
phy_fw_down procedure.
This patch fixes the issue with the link always down on MCBIN board.
nhed: added NULL deref test.
Signed-off-by: Grzegorz Jaszczyk <jaz@semihalf.com>
Reviewed-by: Igal Liberman <igall@marvell.com>
Tested-by: Igal Liberman <igall@marvell.com>
Signed-off-by: Nevo Hed <nhed+github@starry.com>
Reviewed-by: Ramon Fried <rfried.dev@gmail.com>
Acked-by: Joe Hershberger <joe.hershberger@ni.com>
Issue:
- Network stuck if autonegotion fails.
Issue root cause:
- When autonegotiation fails during port open procedure, the packet
processor configuration does not finish and open procedure exits
with error.
- However, this doesn't prevent u-boot network framework from
calling send and receive procedures.
- Using transmit and receive functions of misconfigured packet
processor will cause traffic to get stuck.
Fix:
- Continue packet processor configuration even if autonegotiation
fails. Only error message is triggered in this case.
- Exit transmit and receive functions if there is no PHY link
indication.
- U-boot network framework now calls open procedure again during next
transmit initiation.
Signed-off-by: Stefan Chulski <stefanc@marvell.com>
Reviewed-by: Igal Liberman <igall@marvell.com>
Tested-by: Igal Liberman <igall@marvell.com>
Reviewed-by: Ramon Fried <rfried.dev@gmail.com>
Acked-by: Joe Hershberger <joe.hershberger@ni.com>
The AM654 SoC doesn't allow to disabling RGMII TX internal delay in CPSW2G
MAC. Hence, change CPSW2G interface mode to "rgmii-rxid" - RGMII with
internal RX delay provided by the PHY, the MAC will add an TX delay in this
case.
Signed-off-by: Grygorii Strashko <grygorii.strashko@ti.com>
Acked-by: Joe Hershberger <joe.hershberger@ni.com>
Refactor SGMII configuration to group all settings together and reduce
number of MDIO transactions.
Signed-off-by: Grygorii Strashko <grygorii.strashko@ti.com>
Acked-by: Joe Hershberger <joe.hershberger@ni.com>
