21 Sep, 2017
2 commits
-
After soft reset complete, write mac address immediately will fail.
Add delay to work around this problem.Signed-off-by: rick
-
Add cache inval and flush when rx and tx.
Signed-off-by: rick
19 Sep, 2017
2 commits
-
Update the Micrel KSZ90x1 driver for a live tree.
Signed-off-by: Philipp Tomsich
Acked-by: Joe Hershberger
Reviewed-by: Simon Glass -
Update the Designware Ethernet MAC driver to support a live device
tree.Signed-off-by: Philipp Tomsich
Acked-by: Joe Hershberger
Reviewed-by: Simon Glass
15 Sep, 2017
1 commit
-
All these places seem to inherit the codes from the MMC driver where
a FIXME was put in the comment. However the correct operation after
read should be cache invalidate, not flush.The underlying drivers should be responsible for the cache operation.
Remove these codes completely.Signed-off-by: Bin Meng
Reviewed-by: Stefan Roese
Reviewed-by: York Sun
Reviewed-by: Joe Hershberger
Reviewed-by: Simon Glass
Tested-by: York Sun
11 Sep, 2017
3 commits
-
This patch adds support for RGMII protocol
NXP's LDPAA2 support RGMII protocol. LS1088A is the
first Soc supporting both RGMII and SGMII.Signed-off-by: Prabhakar Kushwaha
Signed-off-by: Amrita Kumari
Signed-off-by: Ashish Kumar
Reviewed-by: York Sun -
LS1088A is compliant with the Layerscape Chassis Generation 3 with
eight ARM v8 Cortex-A53 cores in 2 cluster, CCI-400, one 64-bit DDR4
SDRAM memory controller with ECC, Data path acceleration architecture
2.0 (DPAA2), Ethernet interfaces (SGMIIs, RGMIIs, QSGMIIs, XFIs),
QSPI, IFC, PCIe, SATA, USB, SDXC, DUARTs etc.Signed-off-by: Alison Wang
Signed-off-by: Prabhakar Kushwaha
Signed-off-by: Ashish Kumar
Signed-off-by: Raghav Dogra
Signed-off-by: Shaohui Xie
[YS: Revised commit message]
Reviewed-by: York Sun -
Update MC address calculation as per MC design requirement of address
as least significant 512MB address of MC private allocated memory,
i.e. address should point to end address masked with 512MB offset in
private DRAM block.Signed-off-by: Priyanka Jain
Signed-off-by: Ashish Kumar
[YS: reformatted commit message]
Reviewed-by: York Sun
08 Sep, 2017
2 commits
-
%s/Desriptor/Descriptor/g
Signed-off-by: Heinrich Schuchardt
Acked-by: Joe Hershberger -
The old logic always enabled the TX-delay when the phy-mode was set to
PHY_INTERFACE_MODE_RGMII. With this patch we enable the TX delay for
PHY_INTERFACE_MODE_RGMII_ID and PHY_INTERFACE_MODE_RGMII_TXID and
disable it for PHY_INTERFACE_MODE_RGMII.Based on a similar change made in the Linux Realtek PHY driver
by Martin Blumenstingl .Signed-off-by: Madalin Bucur
Acked-by: Joe Hershberger
Acked-by: York Sun
17 Aug, 2017
1 commit
-
Update pfla02 for setenv changes and PHYLIB/etc migration to Kconfig.
Signed-off-by: Tom Rini
16 Aug, 2017
6 commits
-
We are now using an env_ prefix for environment functions. Rename these
other functions as well, for consistency:getenv_vlan()
getenv_bootm_size()
getenv_bootm_low()
getenv_bootm_mapsize()
env_get_default()Suggested-by: Wolfgang Denk
Signed-off-by: Simon Glass -
Rename this function for consistency with env_get().
Signed-off-by: Simon Glass
-
We are now using an env_ prefix for environment functions. Rename these
two functions for consistency. Also add function comments in common.h.Quite a few places use getenv() in a condition context, provoking a
warning from checkpatch. These are fixed up in this patch also.Suggested-by: Wolfgang Denk
Signed-off-by: Simon Glass -
Rename this function for consistency with env_set().
Signed-off-by: Simon Glass
-
We are now using an env_ prefix for environment functions. Rename these
commonly used functions, for consistency. Also add function comments in
common.h.Suggested-by: Wolfgang Denk
Signed-off-by: Simon Glass -
We are now using an env_ prefix for environment functions. Rename setenv()
for consistency. Also add function comments in common.h.Suggested-by: Wolfgang Denk
Signed-off-by: Simon Glass
15 Aug, 2017
2 commits
-
Even after memory free of phydev, priv is still pointing to the
obsolete address.
So update priv->phydev as NULL after memory free.Signed-off-by: Prabhakar Kushwaha
Signed-off-by: Ashish Kumar
Acked-by: Joe Hershberger -
Signed-off-by: Madalin Bucur
Acked-by: Joe Hershberger
13 Aug, 2017
1 commit
-
The GMAC in the RK3368 once again is identical to the incarnation in
the RK3288 and the RK3399, except for where some of the configuration
and control registers are located in the GRF.This adds the RK3368-specific logic necessary to reuse this driver.
Signed-off-by: Philipp Tomsich
Reviewed-by: Simon Glass
Acked-by: Joe Hershberger
11 Aug, 2017
1 commit
10 Aug, 2017
11 commits
-
Set BM poll size once during priv probe and do not
overwrite it during port probe procedure. Pool is common for
all CP ports.Signed-off-by: Stefan Chulski
Tested-by: iSoC Platform CI
Reviewed-by: Nadav Haklai
Reviewed-by: Igal Liberman
Acked-by: Joe Hershberger
Signed-off-by: Stefan Roese -
TX drain in transmit procedure could cause issues due
to race between drain procedure and transmition of descriptor
between AGGR TXQ and physical TXQ.
TXQ will be cleared before moving to Linux by stop procedure.Signed-off-by: Stefan Chulski
Tested-by: iSoC Platform CI
Reviewed-by: Nadav Haklai
Reviewed-by: Igal Liberman
Acked-by: Joe Hershberger
Signed-off-by: Stefan Roese -
MVPP22 driver support 64 Bit arch and require BM pool
high address configuration.Signed-off-by: Stefan Chulski
Tested-by: iSoC Platform CI
Reviewed-by: Nadav Haklai
Reviewed-by: Igal Liberman
Acked-by: Joe Hershberger
Signed-off-by: Stefan Roese -
Remove IRQ configuration from U-Boot PP driver.
U-Boot don't use interrupts and configuration of IRQ in U-Boot
caused crashes in Linux shared interrupt mode.
Also interrupt use is redundant in RX routine since a single RX
queue is used.Signed-off-by: Stefan Chulski
Tested-by: iSoC Platform CI
Reviewed-by: Nadav Haklai
Reviewed-by: Igal Liberman
Acked-by: Joe Hershberger
Signed-off-by: Stefan Roese -
MBUS driver were replaced by AXI in PPv22 and relevant
only for PPv21.Signed-off-by: Stefan Chulski
Tested-by: iSoC Platform CI
Reviewed-by: Nadav Haklai
Reviewed-by: Igal Liberman
Acked-by: Joe Hershberger
Signed-off-by: Stefan Roese -
U-boot use single physical tx queue with size 16 descriptors.
So aggregated tx queue size should be equal to physical tx queue
and cpu descriptor chunk(number of descriptors delivered from
physical tx queue to aggregated tx queue by one chunk) shouldn't be
larger than physical tx queue.Fix:
Set AGGR_TXQ and CPU_DESC_CHUNK to be 16 descriptors, same as
physical TXQ.Signed-off-by: Stefan Chulski
Tested-by: iSoC Platform CI
Reviewed-by: Nadav Haklai
Reviewed-by: Igal Liberman
Acked-by: Joe Hershberger
Signed-off-by: Stefan Roese -
Issue:
BM counters were overrun by probe that called per Network interface and
caused release of wrong number of buffers during remove procedure.Fix:
Use probe_done and num_ports to call init and remove procedure
once per communication controller.Signed-off-by: Stefan Chulski
Tested-by: iSoC Platform CI
Reviewed-by: Igal Liberman
Acked-by: Joe Hershberger
Signed-off-by: Stefan Roese -
This patch enables padding of packets shorter than 64B in TX(set by default).
Disabling of padding causes crashes on MACCIATO board.Signed-off-by: Stefan Chulski
Tested-by: iSoC Platform CI
Reviewed-by: Igal Liberman
Acked-by: Joe Hershberger
Signed-off-by: Stefan Roese -
A8K marvell SoC has two South Bridge communication controllers(CP0 and CP1).
Each communication controller has packet processor ports and MDIO.
On MACHIATOBin board ports from CP1 are connected to mdio on CP0.Issue:
Wrong base address is assigned to MDIO interface during probe.Fix:
Get MDIO address from PHY handler parent base address.This should be refined in the future when MDIO driver is implemented.
Signed-off-by: Stefan Chulski
Tested-by: iSoC Platform CI
Reviewed-by: Igal Liberman
Acked-by: Joe Hershberger
Signed-off-by: Stefan Roese -
This patch add GPIO configuration support in mvpp2x driver.
Driver will handle 10G SFP gpio reset and SFP TX disable. GPIO pins should
be set in device tree.Signed-off-by: Stefan Chulski
Tested-by: iSoC Platform CI
Reviewed-by: Kostya Porotchkin
Reviewed-by: Igal Liberman
Acked-by: Joe Hershberger
Signed-off-by: Stefan Roese -
In fsl_mc_ldpaa_exit(), in case of mc is booted and dpl is applied,
it should return earlier without executing dpbp_exit().Signed-off-by: Santan Kumar
Acked-by: Priyanka Jain
Acked-by: Yogesh Narayan Gaur
Reviewed-by: York Sun
08 Aug, 2017
8 commits
-
The environment variable "disable_giga" can now be used to disable
1000baseTx on the Micrel's KSZ9031.Signed-off-by: Sebastien Bourdelin
Acked-by: Joe Hershberger -
Added the AFE (Analog Front End) settings for stability to the
Broadcom Cygnus phy. This improves the time take to perform
auto negotiation.Signed-off-by: Arun Parameswaran
Acked-by: Joe Hershberger -
move to Kconfig:
CONFIG_BCM_SF2_ETH
CONFIG_BCM_SF2_ETH_DEFAULT_PORT
CONFIG_BCM_SF2_ETH_GMACAlso modified defconfigs of all platforms that use these configs.
Signed-off-by: Suji Velupillai
Tested-by: Suji Velupillai
Reviewed-by: JD Zheng
Reviewed-by: Scott Branden
Signed-off-by: Steve Rae
Reviewed-by: Tom Rini
Acked-by: Joe Hershberger -
The correct option is PHY_MICREL_KSZ90X1, but some configs still
select the 9021 and 9031 options, which are deprecated.Signed-off-by: Alexandru Gagniuc
Reviewed-by: Philipp Tomsich
Acked-by: Joe Hershberger -
There should be no longer be any ksz9000 users that pick up the PHY
driver from ksz8xxx, so remove ksz9000 remnants from there.Signed-off-by: Alexandru Gagniuc
Reviewed-by: Philipp Tomsich
Acked-by: Joe Hershberger -
The KS8721BL and KSZ9021 PHYs are software-incompatible, yet they
share the same ID. Drivers for bothe PHYs cannot safely coexist, so
the solution was to use #ifdefs to select between the two drivers.As a result KSZ9031, which has a unique ID, is now caught in the
crossfire. Unless CONFIG_PHY_MICREL_KSZ9031 is defined, the KSZ9031
will not function properly, as some essential configuration code is
ifdef'd-out.To prevent such situations, move the KSZ9000 drivers to a separate
file, and place them under a separate Kconfig option. While it is
possible to enable both KSZ8000 and KSZ9000 drivers at the same time,
the assumption is that it is highly unlikely for a system to contain
both a KSZ8000 and a KSZ9000 PHY, and that only one of the drivers
will be enabled at any given time.Signed-off-by: Alexandru Gagniuc
Reviewed-by: Philipp Tomsich
Acked-by: Joe Hershberger -
Signed-off-by: Alexandru Gagniuc
Reviewed-by: Philipp Tomsich
Acked-by: Joe Hershberger -
Don't wait forever.
Pass errors back to the caller.Signed-off-by: Joe Hershberger
Acked-by: Marek Vasut
