mrf24j40: replace magic numbers

This patch replaces some magic numbers with defines for register bits,
mask and shifts.

Signed-off-by: Alexander Aring <alex.aring@gmail.com>
Signed-off-by: Marcel H

mrf24j40: replace magic numbers

This patch replaces some magic numbers with defines for register bits,
mask and shifts.

Signed-off-by: Alexander Aring <alex.aring@gmail.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>

show more ...

mrf24j40: change irq trigger type behaviour

This patch changes the irq trigger type value while calling
devm_request_irq by using IRQF_TRIGGER_LOW when no irq type was given.
Additional we add suppo

mrf24j40: change irq trigger type behaviour

This patch changes the irq trigger type value while calling
devm_request_irq by using IRQF_TRIGGER_LOW when no irq type was given.
Additional we add support for change the irq polarity while hw init if
high level or low level triggered irq type are given.

For rising edge triggered irq's the mrf24j40 can't deal with that, this
races at position of tx completion irq, while the irq is disabled we
readout the irq status registers. This will resets the irq line so other
irq's can occur. Wile readout the irq status register the irq is still
disabled and edge triggered interrupts will be ignored.

Signed-off-by: Alexander Aring <alex.aring@gmail.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>

show more ...

mrf24j40: add promiscuous mode support

This patch adds support for promiscuous mode by setting promiscuous (no
frame filtering), disable automatic ack handling and not filtering
frames where the crc

mrf24j40: add promiscuous mode support

This patch adds support for promiscuous mode by setting promiscuous (no
frame filtering), disable automatic ack handling and not filtering
frames where the crc is invalid.

Reviewed-by: Stefan Schmidt <stefan@osg.samsung.com>
Signed-off-by: Alexander Aring <alex.aring@gmail.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>

show more ...

mrf24j40: add tx power support

This patch supports setting of transmit power for the mrf24j40ma
transceiver only. The mrf24j40mc has some amplifier to change the
transmit power, I am currently not s

mrf24j40: add tx power support

This patch supports setting of transmit power for the mrf24j40ma
transceiver only. The mrf24j40mc has some amplifier to change the
transmit power, I am currently not sure how the mapping for this
amplifier looks like.

Signed-off-by: Alexander Aring <alex.aring@gmail.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>

show more ...

mrf24j40: add cca ed level support

This patch supports handling to set the cca energy detection level for
the mrf24j40 transceiver.

Reviewed-by: Stefan Schmidt <stefan@osg.samsung.com>
Signed-off-b

mrf24j40: add cca ed level support

This patch supports handling to set the cca energy detection level for
the mrf24j40 transceiver.

Reviewed-by: Stefan Schmidt <stefan@osg.samsung.com>
Signed-off-by: Alexander Aring <alex.aring@gmail.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>

show more ...

mrf24j40: add cca mode support

This patch supports cca mode handling for mrf24j40.

Reviewed-by: Stefan Schmidt <stefan@osg.samsung.com>
Signed-off-by: Alexander Aring <alex.aring@gmail.com>
Signed-

mrf24j40: add cca mode support

This patch supports cca mode handling for mrf24j40.

Reviewed-by: Stefan Schmidt <stefan@osg.samsung.com>
Signed-off-by: Alexander Aring <alex.aring@gmail.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>

show more ...

mrf24j40: add csma params support

This patch adds supports to change the CSMA parameters. The datasheet
doesn't say anything about max_be value. Seems not configurable and we
assume the 802.15.4 def

mrf24j40: add csma params support

This patch adds supports to change the CSMA parameters. The datasheet
doesn't say anything about max_be value. Seems not configurable and we
assume the 802.15.4 default. But this value must exists because there is
a min_be value.

Signed-off-by: Alexander Aring <alex.aring@gmail.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>

show more ...

mrf24j40: async interrupt handling

This patch removes the threaded irq handling and do a hardirq instead.
We need to switch to spi_async for this step for getting the irq status
register.

Reviewed-

mrf24j40: async interrupt handling

This patch removes the threaded irq handling and do a hardirq instead.
We need to switch to spi_async for this step for getting the irq status
register.

Reviewed-by: Stefan Schmidt <stefan@osg.samsung.com>
Signed-off-by: Alexander Aring <alex.aring@gmail.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>

show more ...

mrf24j40: rework rx handling to async rx handling

This patch prepares that we can do the receive handling inside interrupt
context by using spi_async. This is necessary for introduce a
non-threaded

mrf24j40: rework rx handling to async rx handling

This patch prepares that we can do the receive handling inside interrupt
context by using spi_async. This is necessary for introduce a
non-threaded irq handling.

Also we drop the bit setting for "RXDECINV" at register "BBREG1", we do
a driectly full write of register "BBREG1", because it contains the bit
RXDECINV only.

Reviewed-by: Stefan Schmidt <stefan@osg.samsung.com>
Signed-off-by: Alexander Aring <alex.aring@gmail.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>

show more ...

mrf24j40: rework tx handling to async tx handling

This patch reworks the current transmit API to spi_async handling. We
removed the error handling check, because mac802154 has no chance to
report it

mrf24j40: rework tx handling to async tx handling

This patch reworks the current transmit API to spi_async handling. We
removed the error handling check, because mac802154 has no chance to
report it. Also the transmit timeout handling can't be handled by xmit
async handling, for this usecase we need to implement the netdev
watchdog. These are all unlikely cases which we drop now and should be
provided by netdev watchdog.

We also drop the bit setting for TXNACKREQ at register TXNCON, this is
not necessary. The TXNCON register should set only once for each frame,
previous settings doesn't matter.

Reviewed-by: Stefan Schmidt <stefan@osg.samsung.com>
Signed-off-by: Alexander Aring <alex.aring@gmail.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>

show more ...

mrf24j40: change to frame delivery with crc

This patch changes the frame delivery to mac802154 with crc. This is
useful for monitor interface types which deliver the crc to userspace.

Reviewed-by:

mrf24j40: change to frame delivery with crc

This patch changes the frame delivery to mac802154 with crc. This is
useful for monitor interface types which deliver the crc to userspace.

Reviewed-by: Stefan Schmidt <stefan@osg.samsung.com>
Signed-off-by: Alexander Aring <alex.aring@gmail.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>

show more ...

mrf24j40: use regmap for register access

This patch uses the regmap functions for transceiver register settings
where it's possible. This means everything except the hotpaths like
receive/transmit h

mrf24j40: use regmap for register access

This patch uses the regmap functions for transceiver register settings
where it's possible. This means everything except the hotpaths like
receive/transmit handling.

Reviewed-by: Stefan Schmidt <stefan@osg.samsung.com>
Signed-off-by: Alexander Aring <alex.aring@gmail.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>

show more ...

mrf24j40: add regmap support

This patch introduce regmap support for short and long address space of
mrf24j40. It's only possible to use regmap_read/write/update_bits for
long address range. This is

mrf24j40: add regmap support

This patch introduce regmap support for short and long address space of
mrf24j40. It's only possible to use regmap_read/write/update_bits for
long address range. This is because I added lowlevel bus operation
because the write operation need to set the 12th bit to mark a register
write, but regmap only supports to set bits for register write access in
the first byte. We use other regmap register functions than
read/write/update_bits, so this should be fine.

Reviewed-by: Stefan Schmidt <stefan@osg.samsung.com>
Signed-off-by: Alexander Aring <alex.aring@gmail.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>

show more ...

mrf24j40: add more register defines

For supporting regmap, this patch will add more register defines to
prepare a full register dump by regmap debugfs.

Reviewed-by: Stefan Schmidt <stefan@osg.samsu

mrf24j40: add more register defines

For supporting regmap, this patch will add more register defines to
prepare a full register dump by regmap debugfs.

Reviewed-by: Stefan Schmidt <stefan@osg.samsung.com>
Signed-off-by: Alexander Aring <alex.aring@gmail.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>

show more ...

mrf24j40: add random extended addr generation

The mrf24j40 has no source to get a permanent extended address. This
patch will add a random generated permanent extended address source.

Reviewed-by:

mrf24j40: add random extended addr generation

The mrf24j40 has no source to get a permanent extended address. This
patch will add a random generated permanent extended address source.

Reviewed-by: Stefan Schmidt <stefan@osg.samsung.com>
Signed-off-by: Alexander Aring <alex.aring@gmail.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>

show more ...

mrf24j40: add default channel setting

Per default mrf24j40 has the channel 11 after reset. This patch adds the
right phy default value for the channel setting.

Reviewed-by: Stefan Schmidt <stefan@o

mrf24j40: add default channel setting

Per default mrf24j40 has the channel 11 after reset. This patch adds the
right phy default value for the channel setting.

Reviewed-by: Stefan Schmidt <stefan@osg.samsung.com>
Signed-off-by: Alexander Aring <alex.aring@gmail.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>

show more ...

mrf24j40: add device-tree support

This patch adds devicetree support to mrf24j40 with proper devicetree
compatible strings.

Reviewed-by: Stefan Schmidt <stefan@osg.samsung.com>
Signed-off-by: Alexa

mrf24j40: add device-tree support

This patch adds devicetree support to mrf24j40 with proper devicetree
compatible strings.

Reviewed-by: Stefan Schmidt <stefan@osg.samsung.com>
Signed-off-by: Alexander Aring <alex.aring@gmail.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>

show more ...

mrf24j40: remove spi settings overwrite

This patch removes spi settings while mrf24j40 probing. These settings
cannot be overwrite while device probing where spi controller should be
already configu

mrf24j40: remove spi settings overwrite

This patch removes spi settings while mrf24j40 probing. These settings
cannot be overwrite while device probing where spi controller should be
already configured. These settings need to be setup by device tree or
platform data.

Reviewed-by: Stefan Schmidt <stefan@osg.samsung.com>
Signed-off-by: Alexander Aring <alex.aring@gmail.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>

show more ...

mrf24j40: calling ieee802154_register_hw at last

The function ieee802154_register_hw should always called at last.
Currently we do hardware init and such things after register hardware
into the subs

mrf24j40: calling ieee802154_register_hw at last

The function ieee802154_register_hw should always called at last.
Currently we do hardware init and such things after register hardware
into the subsystem. It could be that the subsystem already call driver
operations while running hardware init.

Reviewed-by: Stefan Schmidt <stefan@osg.samsung.com>
Signed-off-by: Alexander Aring <alex.aring@gmail.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>

show more ...

mrf24j40: use ieee802154_alloc_hw for private data

This patch removes the own private dataroom allocation by calling
devm_kzalloc for devrec and assign this pointer to "devrec->hw->priv".
Instead we

mrf24j40: use ieee802154_alloc_hw for private data

This patch removes the own private dataroom allocation by calling
devm_kzalloc for devrec and assign this pointer to "devrec->hw->priv".
Instead we using like all other drivers ieee802154_alloc_hw and give the
size for the private driver dataroom at the first argument.

Reviewed-by: Stefan Schmidt <stefan@osg.samsung.com>
Signed-off-by: Alexander Aring <alex.aring@gmail.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>

show more ...

mrf24j40: cleanup define identation

This patch replaces the spaces after define by a tab.

Reviewed-by: Stefan Schmidt <stefan@osg.samsung.com>
Signed-off-by: Alexander Aring <alex.aring@gmail.com>

mrf24j40: cleanup define identation

This patch replaces the spaces after define by a tab.

Reviewed-by: Stefan Schmidt <stefan@osg.samsung.com>
Signed-off-by: Alexander Aring <alex.aring@gmail.com>
Signed-off-by: Marcel Holtmann <marcel@holtmann.org>

show more ...

Merge tag 'v4.3-rc1' into spi-fix-doc

Linux 4.3-rc1

Merge branch 'fixes-rc1' into omap-for-v4.3/fixes

Merge tag 'v4.3-rc1' into MTD -next development

Merge tag 'v4.3-rc1' into patchwork

Linux 4.3-rc1

* tag 'v4.3-rc1': (11838 commits)
Linux 4.3-rc1
blk: rq_data_dir() should not return a boolean
writeback: plug writeback in wb_writeback() an

Merge tag 'v4.3-rc1' into patchwork

Linux 4.3-rc1

* tag 'v4.3-rc1': (11838 commits)
Linux 4.3-rc1
blk: rq_data_dir() should not return a boolean
writeback: plug writeback in wb_writeback() and writeback_inodes_wb()
thermal: fix intel PCH thermal driver mismerge
ARCv2: [axs103_smp] Reduce clk for SMP FPGA configs
revert "ocfs2/dlm: use list_for_each_entry instead of list_for_each"
mm/early_ioremap: add explicit #include of asm/early_ioremap.h
fs/seq_file: convert int seq_vprint/seq_printf/etc... returns to void
selftests: enhance membarrier syscall test
selftests: add membarrier syscall test
sys_membarrier(): system-wide memory barrier (generic, x86)
MODSIGN: fix a compilation warning in extract-cert
Revert "writeback: plug writeback at a high level"
scsi_dh: fix randconfig build error
target: use stringify.h instead of own definition
target/user: Fix UFLAG_UNKNOWN_OP handling
target: Remove no-op conditional
target/user: Remove unused variable
target: Fix max_cmd_sn increment w/o cmdsn mutex regressions
target: Attach EXTENDED_COPY local I/O descriptors to xcopy_pt_sess
...

show more ...