Added a new function kwgbe_write_hwaddr for programming egiga
controller's hardware address.
This function will be called for each egiga port being used
Signed-off-by: Prafulla Wadaskar <prafulla@marvell.com>
Signed-off-by: Ben Warren <biggerbadderben@gmail.com>
This driver supports the Altera triple speeds 10/100/1000 ethernet
mac.
Signed-off-by: Thomas Chou <thomas@wytron.com.tw>
Signed-off-by: Ben Warren <biggerbadderben@gmail.com>
This patch ports the opencore 10/100 ethernet mac driver ethoc.c
from linux kernel to u-boot.
Signed-off-by: Thomas Chou <thomas@wytron.com.tw>
Signed-off-by: Ben Warren <biggerbadderben@gmail.com>
SMSC911x chips have alignment function to allow frame payload data
(which comes after 14-bytes ethernet header) to be aligned at some
boundary when reading it from fifo (usually - 4 bytes boundary).
This is done by inserting fake zeros bytes BEFORE actual frame data when
reading from SMSC's fifo.
This function controlled by RX_CFG register. There are bits that
represents amount of fake bytes to be inserted.
Linux uses alignment of 4 bytes. Ethernet frame header is 14 bytes long,
so we need to add 2 fake bytes to get payload data aligned at 4-bytes
boundary.
Linux driver does this by adding IP_ALIGNMENT constant (defined at
skb.h) when calculating fifo data length. All network subsystem of Linux
uses this constant too when calculating different offsets.
But u-boot does not use any packet data alignment, so we don't need to
add anything when calculating fifo data length.
Moreover, driver zeros the RX_CFG register just one line up, so chip
does not insert any fake data at the beginig. So calculated data length
is always bigger by 1 word.
It seems that at almost every packet read we get an underflow condition
at fifo and possible corruption of data. Especially at continuous
transfers, such as tftp.
Just after removing this magic addition, I've got tftp transfer speed as
it aught to be at 100Mbps. It was really slow before.
It seems that fifo underflow occurs only when using byte packing on
32-bit blackfin bus (may be because of very small delay between reads).
Signed-off-by: Valentin Yakovenkov <yakovenkov@niistt.ru>
Signed-off-by: Ben Warren <biggerbadderben@gmail.com>
Cosmetic changes: Few comments updated
Functionality: Rx packet frame size is programming should
be done when port is in disabled state. this is corrected
Signed-off-by: Prafulla Wadaskar <prafulla@marvell.com>
Signed-off-by: Ben Warren <biggerbadderben@gmail.com>
Fix MX27 FEC logic to check validity of the MAC address in fuse.
Only null (empty fuse) or invalid MAC address was retrieved from mx27 fuses before this change.
Signed-off-by: Eric Jarrige <jorasse@armadeus.org>
Signed-off-by: Ben Warren <biggerbadderben@gmail.com>
When gracefully stopping the controller, the driver was continuing if
*either* RX or TX had stopped. We need to wait for both, or the
controller could get into an invalid state.
Signed-off-by: Andy Fleming <afleming@freescale.com>
Signed-off-by: Ben Warren <biggerbadderben@gmail.com>
The current dm9000x driver accesses its memory mapped registers directly
instead of using the standard I/O accessors. This can cause problems on
Blackfin systems as the accesses can get out of order. So convert the
direct volatile dereferences to use the normal in/out macros.
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
Signed-off-by: Ben Warren <biggerbadderben@gmail.com>
This saves the autonegotation delay when not using ethernet in U-Boot
Signed-off-by: Detlev Zundel <dzu@denx.de>
Signed-off-by: Ben Warren <biggerbadderben@gmail.com>
Avoid using the internal eeprom on MX25 like MX51 already does.
Signed-off-by: John Rigby <jcrigby@gmail.com>
Signed-off-by: Ben Warren <biggerbadderben@gmail.com>
For code archeologists, this is a nice example of copy and paste history.
Signed-off-by: Detlev Zundel <dzu@denx.de>
Signed-off-by: Ben Warren <biggerbadderben@gmail.com>
This patch fixes following build warnings for kirkwood_egiga.c
kirkwood_egiga.c: In function "kwgbe_init":
kirkwood_egiga.c:448: warning: dereferencing type-punned pointer will break strict-aliasing rules
kirkwood_egiga.c: In function "kwgbe_recv":
kirkwood_egiga.c:609: warning: dereferencing type-punned pointer will break strict-aliasing rules
Signed-off-by: Prafulla Wadaskar <prafulla@marvell.com>
Signed-off-by: Ben Warren <biggerbadderben@gmail.com>
Only fill the device enetaddr with the contents of the eeprom,
do not program it in MAC address registers
Signed-off-by: Heiko Schocher <hs@denx.de>
Acked-by: Ben Warren <biggerbadderben@gmail.com>
This code has compile problems and the company does not even exist any
more. So we take the liberty to drop support for it.
Signed-off-by: Detlev Zundel <dzu@denx.de>
CC: Wolfgang Denk <wd@denx.de>
CC: Ben Warren <biggerbadderben@gmail.com>
general cleanup
move clock init to cpu_eth_init in cpu/arm926ejs/mx27/generic.c
make MX27 specific phy init conditional on CONFIG_MX27
replace call to imx_get_ahbclk with one to imx_get_fecclk
and define imx_get_fecclk in include/asm-arm/arch-mx27/clock.h
Signed-off-by: John Rigby <jcrigby@gmail.com>
CC: Ben Warren <biggerbadderben@gmail.com>
CC: Fred Fan <fanyefeng@gmail.com>
CC: Tom <Tom.Rix@windriver.com>
The patch add support for the Freescale mx51 processor
to the FEC ethernet driver.
Signed-off-by: Stefano Babic <sbabic@denx.de>
Signed-off-by: Ben Warren <biggerbadderben@gmail.com>
Although the datasheet mentions seperate smi registers for each
port, using Port 1 smi register to access ethernet phys does not
work. Hence only Port 0 smi register should be used to access all
devices connected to the smi bus. This behavior is consistant with
the mv643xx driver in the linux kernel.
Signed-off-by: Siddarth Gore <gores@marvell.com>
Acked-by: Prafulla Wadaskar <prafulla@marvell.com>
Signed-off-by: Ben Warren <biggerbadderben@gmail.com>
* add's at91_emac (AT91RM9200) network driver (NET_MULTI api)
* enable driver with CONFIG_DRIVER_AT91EMAC
* generic PHY initialization
* modify AT91RM9200 boards to use NET_MULTI driver
* the drivers has been tested with LXT971 Phy and DM9161 Phy at
MII and RMII interface
Signed-off-by: Jens Scharsig <js_at_ng@scharsoft.de>
Signed-off-by: Ben Warren <biggerbadderben@gmail.com>
The EMAC IP on DM365, DM646x and DA830 is slightly different
from that on DM644x. This change updates the DaVinci EMAC driver
so that EMAC becomes operational on SOCs with EMAC v2.
Signed-off-by: Nick Thompson <nick.thompson@ge.com>
Signed-off-by: Ben Warren <biggerbadderben@gmail.com>
Enabling CONFIG_CMD_MII in AVR32 boards was not possible due to
compile errors.
This patch fixes miiphy_read and miiphy_write functions and
registers them properly.
Signed-off-by: Semih Hazar <semih.hazar@indefia.com>
Signed-off-by: Ben Warren <biggerbadderben@gmail.com>
The TSEC_FIBER flag should be set when a PHY is operating with an
external fiber interface. Currently it is only used to notify a user
that the PHY is operating in fiber mode.
A short description was also added to the other TSEC flag defines so
that it is clear how they differ from one another.
Signed-off-by: Peter Tyser <ptyser@xes-inc.com>
Signed-off-by: Ben Warren <biggerbadderben@gmail.com>
The BCM5482 PHY supports both copper and fiber as an ethernet medium.
By enabling its copper/fiber mode auto-detection feature it can
dynamically determine if it should be configured for copper or fiber.
Signed-off-by: Peter Tyser <ptyser@xes-inc.com>
Signed-off-by: Ben Warren <biggerbadderben@gmail.com>
This is generally good practice and saves ~150 bytes.
Signed-off-by: Peter Tyser <ptyser@xes-inc.com>
Signed-off-by: Ben Warren <biggerbadderben@gmail.com>
- Remove unnecessary printing "Enet starting in <speed>/<duplex>"
This same information is already printed during normal ethernet
operation in the form "Speed: 1000, full duplex".
- Add a check for link before determining link speed and duplex
If there is no link, speed/duplex don't matter. This also removes
the annoying and unneeded "Auto-neg error, defaulting to 10BT/HD"
message that occurs when no link is detected.
- Whitespace and line > 80 characters cleanup
Signed-off-by: Peter Tyser <ptyser@xes-inc.com>
Signed-off-by: Ben Warren <biggerbadderben@gmail.com>
In SGMII mode the link between a processor's internal TBI PHY and an
external PHY should always be 1000Mbps, full duplex. Also, the SGMII
interface between an internal TBI PHY and external PHY does not
support in-band auto-negotation.
Previously, when configured for SGMII mode a TBI PHY would attempt to
restart auto-negotation during initializtion. This auto-negotation
between a TBI PHY and external PHY would fail and result in unusable
ethernet operation.
Forcing the TBI PHY and and external PHY to link at 1000Mbps full duplex
in SGMII mode resolves this issue of auto-negotation failing.
Note that 10Mbps and 100Mbps operation is still possible on the external
side of the external PHY even when SGMII is operating at 1000Mbps.
The SGMII interface still operates at 1000Mbps, but each byte of data
is repeated 100 or 10 times for 10/100Mbps and the external PHY handles
converting this data stream into proper 10/100Mbps signalling.
Signed-off-by: Peter Tyser <ptyser@xes-inc.com>
Signed-off-by: Ben Warren <biggerbadderben@gmail.com>
This patch turns off MAC address mismatch warning when
optional eeprom programmed with MAC address is not available.
In that case, smc911x's MAC address register has its default
value ff:ff:ff:ff:ff:ff and it's not a valid address. This
makes eth_initialize() show the warning which has no
meaningful information while environment variable ethaddr
overrides the address read from the register. If there's no
eeprom and the value of MAC address register is not valid
after initialization, dev->enetaddr had better not be updated
and maintain its initial value 00:00:00:00:00:00, which I
think is what eth_initialize() expects. This is not a bug fix.
Even without this patch, the driver works fine. It's just for
enhancing the way of displaying messages.
Signed-off-by: Seunghyeon Rhee <seunghyeon@lpmtec.com>
Signed-off-by: Ben Warren <biggerbadderben@gmail.com>
Moved the mdio regs out of the tsec structure,and
provided different offsets for tsec base and mdio
base so that provision for etsec2.0 can be provided.
This patch helps in providing the support for etsec2.0
In etsec2.0, the MDIO register space and the etsec reg
space are different.
Also, moved the TSEC_BASE_ADDR and MDIO_BASE_ADDR definitons into
platform specific files.
Signed-off-by: Sandeep Gopalpet <sandeep.kumar@freescale.com>
Acked-by: Kim Phillips <kim.phillips@freescale.com>
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
commit 60f61e6d76 breaks compile with gcc by introducing __func__
instead of constant string "func" in the macro call but missed to change the macro.
Signed-off-by: Thomas Weber <weber@corscience.de>
Signed-off-by: Ben Warren <biggerbadderben@gmail.com>
fec_init() will only allocate fec->base_ptr if it is non-NULL. But
the cleanup routine on error will free the pointer without setting
it to NULL. This means that a later call to fec_init() would result
in using an invalid pointer.
Signed-off-by: John Ogness <john.ogness@linutronix.de>
Signed-off-by: Ben Warren <biggerbadderben@gmail.com>
This chip is equipped for example on the esd PMC-ETH2-GB board. So let's
add it to the list of supported chips to the e1000 driver.
Signed-off-by: Reinhard Arlt <reinhard.arlt@esd.eu>
Signed-off-by: Stefan Roese <sr@denx.de>
Signed-off-by: Ben Warren <biggerbadderben@gmail.com>
Make the lan91c96 driver capable of CONFIG_NET_MULTI
to be clean for the new arch, add a a lil detect function
Most of the formatting change was done to keep checkpatch
silent, but a few functions and #if 0ed code which
does not make sense for NET_MULTI have been removed
Now, use the lan91c96_initialize() function to init the driver
Signed-off-by: Nishanth Menon <nm@ti.com>
Signed-off-by: Ben Warren <biggerbadderben@gmail.com>
Make smc911x_initialize return -1 on error and number of interfaces
detected otherwise.
Signed-off-by: Mike Rapoport <mike@compulab.co.il>
Acked-by: Mike Frysinger <vapier@gentoo.org>
Signed-off-by: Ben Warren <biggerbadderben@gmail.com>
Sometimes, inside NetLoop, eth_halt() is called before eth_init() has
been called. This is harmless except for free() calls to pointers
which have not been allocated yet.
This patch initializes those pointers to NULL and allocates them only
the first time. This way we can get rid of free calls in halt callback.
This has been tested in i.MX27 Litekit board and eldk-4.2 toolchains.
Signed-off-by: Javier Martin <javier.martin@vista-silicon.com>
Signed-off-by: Ben Warren <biggerbadderben@gmail.com>
This patch fixes erroneous access to the ethernet PHY which broke the driver.
1. Selector field in the auto-negotiation register must be 0x00001 for
using 802.3, not 0x00000 which is reseved.
2. Access to the PHY address specified by CONFIG_FEC_MXC_PHYADDR, not
0x0 fixed address.
This has been tested in i.MX27 Litekit board and eldk-4.2 toolchains.
Now using proper defines for auto-negotiation register.
Signed-off-by: Javier Martin <javier.martin@vista-silicon.com>
Signed-off-by: Ben Warren <biggerbadderben@gmail.com>
This patch fixes the following warnings:
Configuring for xaeniax board...
smc91111_eeprom.c: In function 'print_macaddr':
smc91111_eeprom.c:278: warning: suggest parentheses around + or - in operand of &
smc91111_eeprom.c:281: warning: suggest parentheses around + or - in operand of &
...
Configuring for xsengine board...
smc91111_eeprom.c: In function 'print_macaddr':
smc91111_eeprom.c:278: warning: suggest parentheses around + or - inside shift
smc91111_eeprom.c:281: warning: suggest parentheses around + or - inside shift
Signed-off-by: Ben Warren <biggerbadderben@gmail.com>
The following error was seen on impa7 board, due to its use of a 32-bit bus
on CS8900.
cs8900.c:137:37: error: macro "get_reg_init_bus" passed 2 arguments, but takes just 1
This patch gives the macro the correct number of arguments
Signed-off-by: Ben Warren <biggerbadderben@gmail.com>
Ensure all CS8900 data structures are assigned before accessing device
Signed-off-by: Hui.Tang <zetalabs@gmail.com>
Signed-off-by: Ben Warren <biggerbadderben@gmail.com>
Proper behavior is to pull MAC address from NVRAM in the initialization() an
stuff it in dev->address, then program the device from dev->address in
the init() function.
Signed-off-by: Ben Warren <biggerbadderben@gmail.com>
