Merge branch 'for-next/misc'

Merge branch 'for-next/lseek'

Merge branch 'for-next/imx'

Merge branch 'for-next/e1000'

net: ethernet: cpsw: cpsw-phy-sel: Use phandle for phy sel ...

With the usage of the interconnect target module hierarchy in device tree the
cpsw-phy-sel node moved to the system control module. An phandle has
been added instead to the device tree include.

Try to use the cpsw-phy-sel phandle first then fall back looking for the
cpsw-phy-sel child.

This patch is based on the upstream kernel patch:
18eb8aea7fb2 ("net: ethernet: cpsw-phy-sel: prefer phandle for phy sel")

Signed-off-by: Teresa Remmet <t.remmet@phytec.de>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

net: phy: micrel: backport ksz9031 125MHz ref-clk workaround ...

This patch backports linux commit e1b505a60366 ("net: phy: micrel: add
125MHz reference clock workaround").

8<--------------------- The original commit message -------------------

net: phy: micrel: add 125MHz reference clock workaround

The micrel KSZ9031 phy has a optional clock pin (CLK125_NDO) which can be
used as reference clock for the MAC unit. The clock signal must meet the
RGMII requirements to ensure the correct data transmission between the
MAC and the PHY. The KSZ9031 phy does not fulfill the duty cycle
requirement if the phy is configured as slave. For a complete
describtion look at the errata sheets: DS80000691D or DS80000692D.

The errata sheet recommends to force the phy into master mode whenever
there is a 1000Base-T link-up as work around. Only set the
"micrel,force-master" property if you use the phy reference clock provided
by CLK125_NDO pin as MAC reference clock in your application.

Attenation, this workaround is only usable if the link partner can
be configured to slave mode for 1000Base-T.

Signed-off-by: Markus Niebel <Markus.Niebel@tqs.de>
[m.felsch@pengutronix.de: fix dt-binding documentation]
[m.felsch@pengutronix.de: use already existing result var for read/write]
[m.felsch@pengutronix.de: add error handling]
[m.felsch@pengutronix.de: add more comments]
Signed-off-by: Marco Felsch <m.felsch@pengutronix.de>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
8<---------------------------------------------------------------------

Signed-off-by: Marco Felsch <m.felsch@pengutronix.de>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

net: fec_imx: skip fec_halt if the FEC wasn't started ...

If the FEC driver is enabled, but the ethernet device was never
opened due to booting from another source the wait for graceful
transmit stop will always fail. This introduces an unnecessary
boot delay and prints a distracting warning.

As there is no point in trying to shut down the FEC if it was
never started, just skip all of fec_halt in that case.

Signed-off-by: Lucas Stach <l.stach@pengutronix.de>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

net/e1000: Do not hardcode TDBAH and RDBAH to 0 ...

While there currently no user of e1000 driver that places those rings
beyond 4GiB boundary, there's also no real reason not to initialize
those registers properly.

Signed-off-by: Andrey Smirnov <andrew.smirnov@gmail.com>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

net/e1000: Improve RX buffer handling in e1000_fill_rx() ...

Drop explicit "volatile" specifier for struct e1000_rx_desc, "bla"
variable as well as explicit endiannes fix, by using little endian IO
accessors (readl, writel, etc.)

Signed-off-by: Andrey Smirnov <andrew.smirnov@gmail.com>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

net/e1000: Consolidate next index calculation code ...

Consolidate next index calculation code into a helper function and
convert the code to make use of it.

Signed-off-by: Andrey Smirnov <andrew.smirnov@gmail.com>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

net/e1000: Remove pointer arithmetic from e1000_fill_rx() ...

Instead of relying on reles of pointer arithmetic (implicit
multiplication by the size of pointer type), change the code to
retreive address of an array elemet to clarify the intent.

Signed-off-by: Andrey Smirnov <andrew.smirnov@gmail.com>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

net/e1000: Rename fill_rx() to e1000_fill_rx() ...

Rename fill_rx() to e1000_fill_rx() to match the naming convention of
other functions in the file.

Signed-off-by: Andrey Smirnov <andrew.smirnov@gmail.com>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

net/e1000: Make use of readl_poll_timeout() in e1000_transmit() ...

Simplify code of e1000_transmit() with readl_poll_timeout().

Signed-off-by: Andrey Smirnov <andrew.smirnov@gmail.com>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

net/e1000: Improve Tx descriptor handling in e1000_transmit ...

Drop explicit "volatile" specifier for struct e1000_tx_desc as well as
explicit endiannes fix, by using little endian IO accessors (readl,
writel, etc.)

Signed-off-by: Andrey Smirnov <andrew.smirnov@gmail.com>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

net/e1000: Remove pointer arithmetic in e1000_transmit() ...

Instead of relying on reles of pointer arithmetic (implicit
multiplication by the size of pointer type), change the code to
retreive address of an array elemet to clarify the intent.

Signed-off-by: Andrey Smirnov <andrew.smirnov@gmail.com>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

net/e1000: Improve Rx descriptor handling in e1000_poll() ...

Drop explicit volatile specifier as well as endianness conversion by
changing the code to use appropriate read*() IO accessors.

While at it if fix incorrect width used for "status" (8 vs 32) and
"len" (16 vs 32).

Signed-off-by: Andrey Smirnov <andrew.smirnov@gmail.com>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

net/e1000: Get rid of pointer arithmetic in e1000_poll ...

Instead of relying on reles of pointer arithmetic (implicit
multiplication by the size of pointer type), change the code to
retreive address of an array elemet to clarify the intent.

Signed-off-by: Andrey Smirnov <andrew.smirnov@gmail.com>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

net/e1000: Fix incorrect "Rx ready" check ...

Due to wrong placement of parenthesis in

	if (!(le32_to_cpu(rd->status)) & E1000_RXD_STAT_DD)
		return 0;

instead of checking that E1000_RXD_STAT_DD is not set, the condition
ends up checking that "status" is 0.

Change the code to invert the condition tested and get rid of !
entirely.

Signed-off-by: Andrey Smirnov <andrew.smirnov@gmail.com>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

net/e1000: Fix debug print warning ...

Cast eeprom->word_size to int, so it would match its printf specifier
to avoid getting errors when building on AArch64.

Signed-off-by: Andrey Smirnov <andrew.smirnov@gmail.com>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

net/e1000: Convert e1000_transmit to use dma_map_single() ...

Convert e1000_transmit to use dma_map_single()/dma_unmap_single(),
which is more apporpirate in this case, since it will account for
phys/virtual address difference as well as allow us to check for DMA
mapping failure.

Signed-off-by: Andrey Smirnov <andrew.smirnov@gmail.com>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

net/e1000: Don't use coherent memory for Rx buffer ...

In order to avoid issues on AArch64, convert the driver to use regular
memory and add appropriate DMA sync calls. Drop needless (uchar *)
cast while at it.

Signed-off-by: Andrey Smirnov <andrew.smirnov@gmail.com>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

net: designware: drop bad RX frames ...

In dwc_ether_rx(), check the RX descriptor status for various error
conditions.  On error, issue a warning with the error status bits and
drop the received frame.

Signed-off-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

net: designware: fix TX descriptor length corruption ...

dwc_ether_send() sets the SIZE1 field of the TX descriptor incorrectly.
It sets the SIZE1 field to a bit-wise OR of the old value and the new
value.  If the old value differs from the new value, the transmitted
Ethernet frame will be longer than it should be and is likely to have an
incorrect frame check sequence.  As more and more Ethernet frames of
different lengths are sent, more and more of the TX descriptors are
affected (up to a maximum number of CONFIG_TX_DESCR_NUM, defined as 16
in "designware.h").  Fix it by setting the SIZE1 field of the TX
descriptor to just the new value.

Signed-off-by: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

devfs: Drop dev_lseek_default() ...

Only the following cdevs do not declare an .lseek() operation:

  - Console devices in common/console.c
  - Firmware framework in common/firmware.c
  - JTAG driver in drivers/misc/jtag.c
  - UBI in drivers/mtd/ubi/barebox.c

Of those four, first two are marked DEVFS_IS_CHARACTER_DEV and
implement only .write() operation and the last two don't implement
anything but .ioctl(). While there's probably no meaningful way to use
lseek() against any of those devices, there doesn't seem to be any
harm in allowing it either.

Change devfs_lseek() to ignore absense of .lseek() callback and drop
dev_lseek_default() and all references to it in the codebase.

Signed-off-by: Andrey Smirnov <andrew.smirnov@gmail.com>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>

Merge branch 'for-next/ultrascale'

Merge branch 'for-next/net'

Merge branch 'for-next/missing-prototypes'

Merge branch 'for-next/misc'

Merge branch 'for-next/e1000'

net: ks8851_mll: fix "no previous prototype for 'ks_enable_qmu'" warning ...

The patch fixes the following compiler's warning:

drivers/net/ks8851_mll.c:438:6: warning: no previous prototype for ‘ks_enable_qmu’ [-Wmissing-prototypes]
 void ks_enable_qmu(struct ks_net *ks)
      ^~~~~~~~~~~~~

Signed-off-by: Antony Pavlov <antonynpavlov@gmail.com>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>