GitLab

With the CONFIG_DM_MMC_OPS migration, the .set_ios callback can
return an integer now.  Return an appropriate error value rather
than sudden death by BUG().
Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>

No more reason to define this function above the ops structure.
Move it near the caller.  Also, change its return type to void
because it never fails.
Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>

Catch up with the DM migration.

As struct dm_mmc_ops does not have .init callback, call the init
function directly from the probe function.
Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>

They are both only referenced in this file.
Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>

Simon Baatz authored Aug 10, 2016 and Stefan Roese committed Aug 26, 2016

Commit f4db6c97

 ("arm: mvebu: Add runtime detection of UART (xmodem)
boot-mode") added a change to hdr->destaddr when dynamically patching an
image for UART boot mode.  With this change, kwboot ceases to work on
Kirkwood.

Thus, let's change hdr->destaddr only when we are patching an image with
header version 1 (Orion and Kirkwood use header version 0).
Signed-off-by: Simon Baatz <gmbnomis@gmail.com>
Fixes: f4db6c97

 ("arm: mvebu: Add runtime detection of UART (xmodem) boot-mode")
Cc: Stefan Roese <sr@denx.de>
Cc: Luka Perkov <luka.perkov@sartura.hr>
Cc: Kevin Smith <kevin.smith@elecsyscorp.com>
Signed-off-by: Stefan Roese <sr@denx.de>

Chris Packham authored Aug 23, 2016 and Stefan Roese committed Aug 26, 2016

Signed-off-by: Chris Packham <judge.packham@gmail.com>
Signed-off-by: Stefan Roese <sr@denx.de>

Chris Packham authored Aug 22, 2016 and Stefan Roese committed Aug 26, 2016

Ensure appropriate error messages are generated. Previously all errors
indicated that the serdes was already in use. Now appropriate error
messages are given.
Signed-off-by: Chris Packham <judge.packham@gmail.com>
Signed-off-by: Stefan Roese <sr@denx.de>

Chris Packham authored Aug 22, 2016 and Stefan Roese committed Aug 26, 2016

As of commit 88e34e5f

 ("spl: replace CONFIG_SPL_SPI_* with
CONFIG_SF_DEFAULT_*") these defines are not used. Remove them to avoid
confusion.
Signed-off-by: Chris Packham <judge.packham@gmail.com>
Signed-off-by: Stefan Roese <sr@denx.de>

Chris Packham authored Aug 22, 2016 and Stefan Roese committed Aug 26, 2016

The NAND interface on the Armada-38x series is similar to that on the
Armada-XP. The key difference is that the NAND ECC clock ratio is
provided via the DFX Server registers instead of the Core Clock.
Signed-off-by: Chris Packham <chris.packham@alliedtelesis.co.nz>
Cc: Luka Perkov <luka.perkov@sartura.hr>
Cc: Dirk Eibach <eibach@gdsys.de>
Signed-off-by: Stefan Roese <sr@denx.de>

Stefan Roese authored Aug 25, 2016 and Heiko Schocher committed Aug 26, 2016

Patch f8a10ed1

 [i2c: mvtwsi: Make address length variable] accidentally
inverted the sequence of address bytes sent to the I2C device. This
patch corrects this by sending the highest byte first and the lowest
byte last again.

Tested on theadorable Armada-XP board.
Signed-off-by: Stefan Roese <sr@denx.de>
Cc: Mario Six <mario.six@gdsys.cc>
Cc: Heiko Schocher <hs@denx.de>

Stephen Warren authored Aug 18, 2016 and Tom Warren committed Aug 25, 2016

In tegra20_slink.c, the set_mode() function may be executed before the
SPI bus is claimed the first time, and hence the clocks to the SPI
controller may not be running. If so, any register read/write at this
time will hang the CPU. Fix this by ensuring the clock is running as soon
as the driver is probed. This is observed on the Tegra30 Beaver board.

Apply the same clock initialization fix to all other Tegra SPI drivers so
that if set_mode() is ever implemented there, the same bug will not appear.
Note that tegra114_spi.c already operates in this fashion.

The clock manipulation code is copied from claim_bus() to probe() rather
than moved. This ensures that any calls to set_speed() take effect; the
clock can't be set once during probe and left unchanged.

Fixes: 5cb1b7b3

 ("spi: tegra20: Add support for mode selection")
Cc: Mirza Krak <mirza.krak@hostmobility.com>
Signed-off-by: Stephen Warren <swarren@nvidia.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
Signed-off-by: Tom Warren <twarren@nvidia.com>

Stephen Warren authored Aug 22, 2016 and Tom Warren committed Aug 25, 2016

The nvidia,bpmp property is left over from an old BPMP I2C binding, and
shouldn't be present. Remove it from the SoC DT file, and update the
I2C driver not to parse it; the value wasn't used for anything any more
anyway.
Signed-off-by: Stephen Warren <swarren@nvidia.com>
Signed-off-by: Tom Warren <twarren@nvidia.com>

Stephen Warren authored Aug 18, 2016 and Tom Warren committed Aug 25, 2016

The Tegra SDHCI binding dictates that the reseet name for the Tegra SDHCI
clock be "sdhci" not "sdmmc", and that the clock is accessed by index
rather than by name. Fix the Tegra186 DT and MMC driver to honor this.
Signed-off-by: Stephen Warren <swarren@nvidia.com>
Signed-off-by: Tom Warren <twarren@nvidia.com>

Stephen Warren authored Aug 18, 2016 and Tom Warren committed Aug 25, 2016

The Tegra I2C binding dictates that the clock name for the Tegra I2C clock
be "div-clk" not "i2c". Fix the Tegra186 DT and I2C driver to honor this.
Signed-off-by: Stephen Warren <swarren@nvidia.com>
Signed-off-by: Tom Warren <twarren@nvidia.com>

Signed-off-by: Tom Rini <trini@konsulko.com>

The fs-test.sh script expected there to be a \n\r style newline at the
end of the output. This is no longer the case, so use 'tr' to remove the
\r that we get.

Fixes: (c5917b4b

 "dm: serial-uclass: Move a carriage return before a
        line feed")
Signed-off-by: Tom Rini <trini@konsulko.com>

Dongpo Li authored Aug 22, 2016 and Joe Hershberger committed Aug 22, 2016

When phy autoneg on, the link speed and duplex should be
determined by phy advertising register and
phy link partner ability register.
Check phy advertising register when geting phy link speed and
duplex if autoneg on.
Signed-off-by: Dongpo Li <lidongpo@hisilicon.com>
Cc: Joe Hershberger <joe.hershberger@ni.com>
Acked-by: Joe Hershberger <joe.hershberger@ni.com>

karl beldan authored Aug 20, 2016 and Joe Hershberger committed Aug 22, 2016

The spatch series converting legacy drivers from miiphy_register to
mdio_register changed the return convention of the davinci_emac internal
MDIO accessors, making the internal code relying on it misbehaving:
no mdiodev get registered and U-Boot crashes when using net cmds in the
context of the old legacy net API.

ATM davinci_emac_initialize and cpu_eth_init don't return a proper value
in that case but fixing them would not avoid the crash.

This change is just a follow-up to the spatch pass, the MDIO accessors
of the mdiodev introduced by the spatch pass retain their proper values.
Signed-off-by: Karl Beldan <karl.beldan+oss@gmail.com>
Acked-by: Joe Hershberger <joe.hershberger@ni.com>

Hou Zhiqiang authored Aug 19, 2016 and Joe Hershberger committed Aug 22, 2016

The QMan is not used in FMan IM mode, so no QMI enqueue or QMI
dequeue are performed.
Signed-off-by: Hou Zhiqiang <Zhiqiang.Hou@nxp.com>
Acked-by: Joe Hershberger <joe.hershberger@ni.com>

karl beldan authored Aug 15, 2016 and Joe Hershberger committed Aug 22, 2016

ATM when receiving a packet the whole buffer is invalidated, this change
optimizes this behaviour.
Signed-off-by: Karl Beldan <karl.beldan+oss@gmail.com>
Acked-by: Joe Hershberger <joe.hershberger@ni.com>
Reviewed-by: Tom Rini <trini@konsulko.com>
Reviewed-by: Mugunthan V N <mugunthanvnm@ti.com>

karl beldan authored Aug 15, 2016 and Joe Hershberger committed Aug 22, 2016

check_cache_range() warns that the top boundaries are not properly
aligned when flushing or invalidating the buffers and make these
operations fail.

This gets rid of the remaining warnings:
CACHE: Misaligned operation at range
Signed-off-by: Karl Beldan <karl.beldan+oss@gmail.com>
Acked-by: Joe Hershberger <joe.hershberger@ni.com>
Reviewed-by: Tom Rini <trini@konsulko.com>
Reviewed-by: Mugunthan V N <mugunthanvnm@ti.com>

karl beldan authored Aug 14, 2016 and Joe Hershberger committed Aug 22, 2016

ATM the rx and tx descriptors are handled as cached memory while they
lie in a dedicated RAM of the SoCs, which is an uncached area.
Removing the said dcache ops, while optimizing the logic and clarifying
the code, also gets rid of most of the check_cache_range() incurred
warnings:
CACHE: Misaligned operation at range
Signed-off-by: Karl Beldan <karl.beldan+oss@gmail.com>
Acked-by: Joe Hershberger <joe.hershberger@ni.com>

We use an empty hostname, so remove all the "processing" of the
known-to-be-empty hostname and just write 0's where needed.
Signed-off-by: Joe Hershberger <joe.hershberger@ni.com>

Instead of always allocating a huge temporary buffer on the stack and
then memcpy()ing the result into the transmit buffer, simply figure out
where in the transmit buffer the bytes will belong and write them there
directly as each message is built.
Signed-off-by: Joe Hershberger <joe.hershberger@ni.com>

Much of the information is verbose and derived directly from the
environment. Only output in debug mode. This also saves about 300 bytes
from the code size.
Signed-off-by: Joe Hershberger <joe.hershberger@ni.com>

Use the same name throughout the nfs code and use the same member of the
union to avoid casts.
Signed-off-by: Joe Hershberger <joe.hershberger@ni.com>

The buffer is of 32-bit elements, not bytes.
Signed-off-by: Joe Hershberger <joe.hershberger@ni.com>

Instead of repeating the same large snippet for dealing with attributes
it should be shared with a helper function.
Signed-off-by: Joe Hershberger <joe.hershberger@ni.com>

Fix complaints from checkpatch.pl.
Signed-off-by: Joe Hershberger <joe.hershberger@ni.com>

The type of the buffer is uint32_t, but the parameter used to size it
is referring to bytes. Divide by the size of the array elements.

Strictly speaking, this shouldn't be needed at all... It could just be 1
just like the request.
Signed-off-by: Joe Hershberger <joe.hershberger@ni.com>

The v3 handles can be larger than v2, but that doesn't mean we need a
separate buffer. Reuse the same (larger) buffer for both.
Signed-off-by: Joe Hershberger <joe.hershberger@ni.com>

Guillaume GARDET authored Jul 29, 2016 and Joe Hershberger committed Aug 22, 2016

This patch enables NFSv3 support.
If NFSv2 is available use it as usual.
If NFSv2 is not available, but NFSv3 is available, use NFSv3.
If NFSv2 and NFSv3 are not available, print an error message since NFSv4 is not supported.

Tested on iMX6 sabrelite with 4 Linux NFS servers:
  * NFSv2 + NFSv3 + NFSv4 server: use NFSv2 protocol
  * NFSv2 + NFSv3 server: use NFSv2 protocol
  * NFSv3 + NFSv4 server: use NFSv3 protocol
  * NFSv3 server: use NFSv3 protocol
Signed-off-by: Guillaume GARDET <guillaume.gardet@free.fr>
Cc: Tom Rini <trini@konsulko.com>
Cc: joe.hershberger@ni.com
Acked-by: Joe Hershberger <joe.hershberger@ni.com>

Unreferenced, so remove the noise.
Signed-off-by: Joe Hershberger <joe.hershberger@ni.com>

There is no reason to store the default filename in a separate buffer
only to immediately copy it to the main name buffer. Just write it there
directly and remove the other buffer.
Signed-off-by: Joe Hershberger <joe.hershberger@ni.com>

At least on bfin, this "specimen" is actually allocated in the BSS and
wastes lots of memory in already tight memory conditions.

Also, with the introduction of NFSv3 support, this waste got
substantially larger.

Just remove it. If a board needs a specific different defragment size,
that board can override this setting.
Signed-off-by: Joe Hershberger <joe.hershberger@ni.com>

[trini: Drop CMD_BOOTI as it's now on by default on ARM64]
Signed-off-by: Tom Rini <trini@konsulko.com>

Masahiro Yamada authored Aug 19, 2016 and Tom Rini committed Aug 20, 2016

'-' is never used in function names.
Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>