if status register do never set MXC_CSPICTRL_TC, spi_xchg_single
endless loops. Add a timeout here to prevent endless hang.
Signed-off-by: Heiko Schocher <hs@denx.de>
Cc: Dirk Behme <dirk.behme@gmail.com>
Reviewed-by: Jagannadha Sutradharudu Teki <jaganna@xilinx.com>
This parameter should also be supported.
Signed-off-by: Simon Glass <sjg@chromium.org>
Tested-by: Ajay Kumar <ajaykumar.rs@samsung.com>
Reviewed-by: Jagannadha Sutradharudu Teki <jaganna@xilinx.com>
The SPI transaction delay is supposed to be measured from the end of one
transaction to the start of the next. The code does not work that way, so
fix it.
Signed-off-by: Simon Glass <sjg@chromium.org>
Tested-by: Ajay Kumar <ajaykumar.rs@samsung.com>
Reviewed-by: Jagannadha Sutradharudu Teki <jaganna@xilinx.com>
This patch returns back support for old ep93xx processors family
Signed-off-by: Sergey Kostanbaev <sergey.kostanbaev@gmail.com>
Cc: albert.u.boot@aribaud.net
SPI recieve and transfer code in exynos_spi driver has a logical bug.
We read data in a variable which can hold an integer. Then we assign
this integer 32 bit value to another variable which has data type uchar.
Latter represents a unit of our recieve buffer. Everytime when we write
a value to our recieve buffer we step ahead by 4 units when actually we
wrote to one unit. This results in the loss of 3 bytes out of every 4
bytes recieved. This patch intends to fix this bug.
Signed-off-by: Akshay Saraswat <akshay.s@samsung.com>
Acked-by: Simon Glass <sjg@chromium.org>
Tested-by: Simon Glass <sjg@chromium.org>
Signed-off-by: Minkyu Kang <mk7.kang@samsung.com>
This mirrors the conventions used in other SPI drivers (kirkwood,
davinci, atmel, et al) where the din/dout buffer can be NULL when the
received/transmitted data isn't important. This reduces the need for
allocating additional buffers when write-only/read-only functionality is
needed.
In the din == NULL case, the received data is simply not stored. In the
dout == NULL case, zeroes are transmitted.
Signed-off-by: Andrew Ruder <andrew.ruder@elecsyscorp.com>
Cc: Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com>
Reviewed-by: Jagannadha Sutradharudu Teki <jaganna@xilinx.com>
The patch populates the slave data which will be used by flash driver to
set the flash quad enable bit.
Signed-off-by: Sourav Poddar <sourav.poddar@ti.com>
Replace 80 mircoseconds delay with polling flag ESPI_EV_TXE.
Signed-off-by: Hou Zhiqiang <B48286@freescale.com>
Reviewed-by: York Sun <yorksun@freescale.com>
Currently davinci spi driver supports only bus 0 cs 0.
This patch allows driver to support bus 1 and bus 2 with
configurable number of chip selects. Also defaults are
selected in a way to avoid regression on other platforms
that uses davinci spi driver and has only one spi bus.
Signed-off-by: Rex Chang <rchang@ti.com>
Signed-off-by: Murali Karicheri <m-karicheri2@ti.com>
Reviewed-by: Jagannadha Sutradharudu Teki <jaganna@xilinx.com>
Bulk erase is not happening properly on dra7 due to erase timing constraints,
add a delay so that erase timing constraints are properly met.
Signed-off-by: Sourav Poddar <sourav.poddar@ti.com>
Tested-by: Yebio Mesfin <ymesfin@ti.com>
This renames all the Tegra20 pinmux pins and functions so they have a
prefix which matches the type name.
The entries in tegra20_pingroups[] are all updated to remove the columns
which are no longer used.
All affected code is updated to match.
Signed-off-by: Stephen Warren <swarren@nvidia.com>
Acked-by: Simon Glass <sjg@chromium.org>
Signed-off-by: Tom Warren <twarren@nvidia.com>
Refactor the code a bit to make it better in readability.
Remove the comments because now the intention of the code is pretty clear.
Signed-off-by: Axel Lin <axel.lin@ingics.com>
Reviewed-by: Jagannadha Sutradharudu Teki <jaganna@xilinx.com>
The logic determining SPI "write" transfer completion was faulty. At
certain conditions (e.g. slow SPI clock freq) the transfers were
interrupted before completion. Both EOT and TXS flags of channel
status registeer shall be checked to ensure that all data was
transferred. Tested on AM3359 chip.
Signed-off-by: Vasili Galka <vasili@visionmap.com>
Without this delay, write/read is failing.
Looks like, the WIP always remain set and hence a timeout
occurs leading to the error.
Without this patch, device does not get probed also.
Here is the log.
U-Boot#
U-Boot#
U-Boot#
U-Boot# sf probe 0
SF: Unsupported flash IDs: manuf ff, jedec ffff, ext_jedec ffff
Failed to initialize SPI flash at 0:0
U-Boot# sf probe 0
While with this patch, log is
U-Boot# sf probe 0
SF: Detected MX25L51235F with page size 256 Bytes, erase size 64 KiB, total 64 MiB, mapped at 30000000
U-Boot# sf erase 0 0x80000
SF: 524288 bytes @ 0x0 Erased: OK
U-Boot# mw 81000000 0xdededede 0x40000
U-Boot# sf write 81000000 0 0x40000
SF: 262144 bytes @ 0x0 Written: OK
U-Boot# sf read 82000000 0 0x40000
SF: 262144 bytes @ 0x0 Read: OK
U-Boot# md 0x82000000
Signed-off-by: Sourav Poddar <sourav.poddar@ti.com>
Reviewed-by: Jagannadha Sutradharudu Teki <jaganna@xilinx.com>
SPI_MODE_3 requires clk high when inactive. The SCLK_CTL
field of the config reg was not configured in case of CPOL.
Fix configuration so that SPI_MODE_3 which uses CPOL configures
the clk line to be high in inactive state.
Signed-off-by: Markus Niebel <Markus.Niebel@tqs.de>
Reviewed-by: Jagannadha Sutradharudu Teki <jaganna@xilinx.com>
Timeout calculation should be out of the data loop.
This patch increase spi bandwidth for 30%.
Signed-off-by: Michal Simek <michal.simek@xilinx.com>
Reviewed-by: Jagannadha Sutradharudu Teki <jaganna@xilinx.com>
We have a sh_spi_clear_bit() function, there's no reason not to use it.
Signed-off-by: Axel Lin <axel.lin@ingics.com>
Acked-by: Nobuhiro Iwamatsu <iwamatsu@nigauri.org>
Reviewed-by: Jagannadha Sutradharudu Teki <jaganna@xilinx.com>
The Faraday FTSSP010 is a multi-function controller
which supports I2S/SPI/SSP/AC97/SPDIF. However This
patch implements only the SPI mode.
NOTE:
The DMA and CS/Clock control logic has been altered
since hardware revision 1.19.0. So this patch
would first detects the revision id of the underlying
chip, and then switch to the corresponding software
control routines.
Signed-off-by: Kuo-Jung Su <dantesu@faraday-tech.com>
Signed-off-by: Jagannadha Sutradharudu Teki <jaganna@xilinx.com>
CC: Tom Rini <trini@ti.com>
The RDY bit indicates that a transfer is complete. This needs to be
cleared by SW before every single HW transaction, rather than only
at the start of each SW transaction (those being made up of n HW
transactions).
It seems that earlier HW may have cleared this bit autonomously when
starting a new transfer, and hence this code was not needed in practice.
However, this is generally a good idea in all cases. In Tegra124, the
HW behaviour appears to have changed, and SW must explicitly clear this
bit. Otherwise, SW will believe that transfers have completed when they
have not, and may e.g. read stale data from the RX FIFO.
Signed-off-by: Yen Lin <yelin@nvidia.com>
[swarren, rewrote commit description, unified duplicate RDY clearing code
and moved it right before the start of the HW transaction, unconditionally
exit loop after reading RX data, rather than checking if TX FIFO is empty,
since it is guaranteed to be]
Signed-off-by: Stephen Warren <swarren@nvidia.com>
Reviewed-by: Jagannadha Sutradharudu Teki <jaganna@xilinx.com>
This patch adds a driver for Renesas SoC's Quad SPI bus.
This supports with 8 bits per transfer to use with SPI flash.
Signed-off-by: Kouei Abe <kouei.abe.cp@renesas.com>
Signed-off-by: Nobuhiro Iwamatsu <nobuhiro.iwamatsu.yj@renesas.com>
Signed-off-by: Jagannadha Sutradharudu Teki <jaganna@xilinx.com>
This adds a SPI framework for people to hook up simulated SPI clients.
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
Signed-off-by: Simon Glass <sjg@chromium.org>
This allows us to put the SPI flash chip inside the SPI interface node,
with U-Boot finding the correct bus and chip select automatically.
Signed-off-by: Simon Glass <sjg@chromium.org>
For invalid bus number, current code returns NULL in the default case of
switch-case statements. In additional, pins[bus] is always not NULL because
it is the address of specific row of the two-dimensional array.
Thus this patch removes these unnecessary test.
Signed-off-by: Axel Lin <axel.lin@ingics.com>
Acked-by: Scott Jiang <scott.jiang.linux@gmail.com>
Signed-off-by: Sonic Zhang <sonic.zhang@analog.com>
For invalid bus number, current code returns NULL in the default case of
switch-case statements. In additional, pins[bus] is always not NULL because
it is the address of specific row of the two-dimensional array.
Thus this patch removes these unnecessary test.
Signed-off-by: Axel Lin <axel.lin@ingics.com>
Acked-by: Scott Jiang <scott.jiang.linux@gmail.com>
Signed-off-by: Sonic Zhang <sonic.zhang@analog.com>
Current implementation only supports 8 bit word lengths, even though
omap3 can handle anything between 4 and 32.
Update the spi interface to support changing the SPI word length,
and implement it in omap3_spi driver to support the full range of
possible word lengths.
This implementation is backwards compatible by defaulting to the old
behavior of 8 bit word lengths.
Also, it required a change to the omap3_spi non static I/O functions,
but since they are not used anywhere else, no collateral changes are required.
Cc: Tom Rini <trini@ti.com>
Cc: Jagannadha Sutradharudu Teki <jagannadh.teki@gmail.com>
Cc: Igor Grinberg <grinberg@compulab.co.il>
Signed-off-by: Nikita Kiryanov <nikita@compulab.co.il>
If dout buffer is not 32 bit-aligned or data to transmit is not multiple
of 32 bit the read data pointer is already incremented on single byte reads.
Signed-off-by: Timo Herbrecher <t.herbrecher@gateware.de>
Signed-off-by: Jagannadha Sutradharudu Teki <jaganna@xilinx.com>
Since SPI register access is so expensive, it is worth transferring data
a word at a time if we can. This complicates the driver unfortunately.
Use the byte-swapping feature to avoid having to convert to/from big
endian in software.
This change increases speed from about 2MB/s to about 4.5MB/s.
Signed-off-by: Simon Glass <sjg@chromium.org>
Signed-off-by: Rajeshwari S Shinde <rajeshwari.s@samsung.com>
Reviewed-by: Jagannadha Sutradharudu Teki <jagannadh.teki@gmail.com>
Accessing SPI registers is slow, but access to the FIFO level register
in particular seems to be extraordinarily expensive (I measure up to
600ns). Perhaps it is required to synchronise with the SPI byte output
logic which might run at 1/8th of the 40MHz SPI speed (just a guess).
Reduce access to this register by filling up and emptying FIFOs
more completely, rather than just one word each time around the inner
loop.
Since the rxfifo value will now likely be much greater that what we read
before we fill the txfifo, we only fill the txfifo halfway. This is
because if the txfifo is empty, but the rxfifo has data in it, then writing
too much data to the txfifo may overflow the rxfifo as data arrives.
This speeds up SPI flash reading from about 1MB/s to about 2MB/s on snow.
Signed-off-by: Simon Glass <sjg@chromium.org>
Signed-off-by: Rajeshwari S Shinde <rajeshwari.s@samsung.com>
Reviewed-by: Jagannadha Sutradharudu Teki <jagannadh.teki@gmail.com>
For devices that need some time to react after a spi transaction
finishes, add the ability to set a delay.
Implement this as a delay on the first/next transaction to avoid
any delay in the fairly common case where a SPI transaction is
followed by other processing.
Signed-off-by: Simon Glass <sjg@chromium.org>
Signed-off-by: Rajeshwari S Shinde <rajeshwari.s@samsung.com>
Reviewed-by: Jagannadha Sutradharudu Teki <jagannadh.teki@gmail.com>
The chipselect must be written into the CTRL0 register after the SSP
block is reset, otherwise the block will always use ChipSelect #0.
Signed-off-by: Marek Vasut <marex@denx.de>
Cc: Fabio Estevam <fabio.estevam@freescale.com>
Cc: Jagannadha Sutradharudu Teki <jagannadh.teki@gmail.com>
Cc: Otavio Salvador <otavio@ossystems.com.br>
Cc: Stefano Babic <sbabic@denx.de>
Acked-by: Stefano Babic <sbabic@denx.de>
85xx, 86xx PowerPC folders have code variables with CamelCase naming conventions.
because of this code checkpatch script generates "WARNING: Avoid CamelCase".
Convert variables name to normal naming convention and modify board, driver
files with updated the new structure.
Signed-off-by: Prabhakar Kushwaha <prabhakar@freescale.com>
Acked-by: York Sun <yorksun@freescale.com>
Use DIV_ROUND_UP to simplify the code.
Signed-off-by: Axel Lin <axel.lin@ingics.com>
Signed-off-by: Scott Jiang <scott.jiang.linux@gmail.com>
Signed-off-by: Sonic Zhang <sonic.zhang@analog.com>
The spi clock divisor is of the form x * (2**y), or x << y, where x is
1 to 16, and y is 0 to 15. Note the similarity with floating point numbers.
Convert the desired divisor to the smallest number which is >= desired divisor,
and can be represented in this form. The previous algorithm chose a divisor
which could be almost twice as large as needed.
Signed-off-by: Troy Kisky <troy.kisky@boundarydevices.com>
Signed-off-by: Dirk Behme <dirk.behme@gmail.com>
Fix two issues with the calculation of pre_div and post_div:
1. pre_div: While the calculation of pre_div looks correct, to set the
CONREG[15-12] bits pre_div needs to be decremented by 1:
The i.MX 6Dual/6Quad Applications Processor Reference Manual (IMX6DQRM
Rev. 0, 11/2012) states:
CONREG[15-12]: PRE_DIVIDER
0000 Divide by 1
0001 Divide by 2
0010 Divide by 3
...
1101 Divide by 14
1110 Divide by 15
1111 Divide by 16
I.e. if we want to divide by 2, we have to write 1 to CONREG[15-12].
2. In case the post divider becomes necessary, pre_div will be divided by
16. So set pre_div to 16, too. And not 15.
Both issues above are tested using the following examples:
clk_src = 60000000 (60MHz, default i.MX6 ECSPI clock)
a) max_hz == 23000000 (23MHz, max i.MX6 ECSPI read clock)
-> pre_div = 3 (divide by 3 => CONREG[15-12] == 2)
-> post_div = 0 (divide by 1 => CONREG[11- 8] == 0)
=> 60MHz / 3 = 20MHz SPI clock
b) max_hz == 2000000 (2MHz)
-> pre_div = 16 (divide by 16 => CONREG[15-12] == 15)
-> post_div = 1 (divide by 2 => CONREG[11- 8] == 1)
=> 60MHz / 32 = 1.875MHz SPI clock
c) max_hz == 1000000 (1MHz)
-> pre_div = 16 (divide by 16 => CONREG[15-12] == 15)
-> post_div = 2 (divide by 4 => CONREG[11- 8] == 2)
=> 60MHz / 64 = 937.5kHz SPI clock
d) max_hz == 500000 (500kHz)
-> pre_div = 16 (divide by 16 => CONREG[15-12] == 15)
-> post_div = 3 (divide by 8 => CONREG[11- 8] == 3)
=> 60MHz / 128 = 468.75kHz SPI clock
Signed-off-by: Dirk Behme <dirk.behme@gmail.com>
Heiko Schocher
Simon Glass
Sergey Kostanbaev
Akshay Saraswat
Axel Lin
Alison Wang
Andrew Ruder
Sourav Poddar
Hou Zhiqiang
Karicheri, Muralidharan
Stephen Warren
Vasili Galka
Sonic Zhang
Markus Niebel
Michal Simek
Nobuhiro Iwamatsu
Kuo-Jung Su
Yen Lin
Mike Frysinger
Nikita Kiryanov
Masahiro Yamada
Timo Herbrecher
Jagannadha Sutradharudu Teki
Wolfgang Denk
Rajeshwari Shinde
Matt Porter
Marek Vasut
Prabhakar Kushwaha
Dirk Behme