[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <ede8544f-c71d-0901-5522-e97215d23f85@st.com>
Date: Mon, 10 Dec 2018 12:37:56 +0000
From: Amelie DELAUNAY <amelie.delaunay@...com>
To: "cezary.gapinski@...il.com" <cezary.gapinski@...il.com>,
Mark Brown <broonie@...nel.org>,
"linux-spi@...r.kernel.org" <linux-spi@...r.kernel.org>,
"linux-stm32@...md-mailman.stormreply.com"
<linux-stm32@...md-mailman.stormreply.com>,
"linux-arm-kernel@...ts.infradead.org"
<linux-arm-kernel@...ts.infradead.org>,
"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
Rob Herring <robh+dt@...nel.org>,
"devicetree@...r.kernel.org" <devicetree@...r.kernel.org>
CC: Maxime Coquelin <mcoquelin.stm32@...il.com>,
Alexandre TORGUE <alexandre.torgue@...com>,
Mark Rutland <mark.rutland@....com>
Subject: Re: [PATCH 0/5] Add support for STM32F4 SPI
Hi Cezary,
On 12/9/18 2:53 PM, cezary.gapinski@...il.com wrote:
> From: Cezary Gapinski <cezary.gapinski@...il.com>
>
> This series of patches adds support for first generation of SPI interface
> for STM32F4 family.
>
First of all, thanks for adding STM32F4 SPI support.
> This version of driver is mostly different to STM32H7 one. Based on linux
> kernel I2C drivers for STM32 where drivers were splited into STM32F4 and
> STM32F7 family the same approach seems to be sufficient for SPI STM32
> drivers. Therefore STM32H7 driver was moved to spi-stm32h7.c file and
> register and functions were renamed to be more specific to STM32H7.
>
You're right, STM32F4 SPI is slightly different from STM32H7 one:
register map/bits offsets are different and STM32H7 has an RX and TX FIFO.
But if you have a look on STM32F7 SPI [1], you'll see that STM32F7 SPI
is based on STM32F4 SPI with new features (data frames & FIFOs) also
available on STM32H7 SPI.
That's why STM32H7 SPI driver was called spi-stm32. The goal was to use
compatible & match data to differentiate each STM32Fx specificities.
You can have a look on how it is managed in drivers/rtc/rtc-stm32.c (the
same driver covers 2 HW version of STM32 RTC), or in
drivers/iio/dac/stm32-dac-core.c and stm32-dac.c (the same driver also
covers 2 HW version of STM32 DAC).
As your spi-stm32f4.c file is highly based on the existing spi-stm32.c
file, I think that common code could be factored and specificities could
be handled with compatible and match data.
Regards,
Amelie
[1]
https://www.st.com/content/ccc/resource/technical/document/reference_manual/c5/cf/ef/52/c0/f1/4b/fa/DM00124865.pdf/files/DM00124865.pdf/jcr:content/translations/en.DM00124865.pdf
> For current version master mode with full-duplex and 8/16 bit data frame
> format are supported. There is no TX and RX FIFOs like in STM32H7.
> DMA capabilility is supported for messages longer than arbitrary number
> of bytes (that is set already to 16 bytes) when TX and RX channels are
> set at the same time.
>
> Cezary Gapinski (5):
> spi: stm32: rename STM32 SPI registers and functions to STM32H7
> spi: stm32: rename spi-stm32 to spi-stm32h7
> spi: stm32: add driver for STM32F4 controller
> ARM: dts: stm32: add SPI support on STM32F429 SoC
> spi: stm32: add description about STM32F4 bindings
>
> .../devicetree/bindings/spi/spi-stm32.txt | 9 +-
> arch/arm/boot/dts/stm32f429.dtsi | 60 +
> drivers/spi/Kconfig | 18 +-
> drivers/spi/Makefile | 3 +-
> drivers/spi/spi-stm32.c | 1322 -------------------
> drivers/spi/spi-stm32f4.c | 1002 +++++++++++++++
> drivers/spi/spi-stm32h7.c | 1340 ++++++++++++++++++++
> 7 files changed, 2424 insertions(+), 1330 deletions(-)
> delete mode 100644 drivers/spi/spi-stm32.c
> create mode 100644 drivers/spi/spi-stm32f4.c
> create mode 100644 drivers/spi/spi-stm32h7.c
>
Powered by blists - more mailing lists