On Thu, 18 Nov 2021 13:32:18 +0100
Ahmad Fatoum <a.fatoum@xxxxxxxxxxxxxx> wrote:
Hi Ahmad!
> On 08.11.21 18:01, Antony Pavlov wrote:
> > On Fri, 5 Nov 2021 08:46:57 +0100
> > Ahmad Fatoum <a.fatoum@xxxxxxxxxxxxxx> wrote:
> >
> >> Tested on a STM32MP1 communicating with a ksz9563.
> >
> > ...
> >
> >> diff --git a/drivers/spi/stm32_spi.c b/drivers/spi/stm32_spi.c
> >> new file mode 100644
> >> index 000000000000..0cb04a968c8a
> >> --- /dev/null
> >> +++ b/drivers/spi/stm32_spi.c
> >> @@ -0,0 +1,590 @@
> >> +// SPDX-License-Identifier: GPL-2.0+ OR BSD-3-Clause
> >> +/*
> >> + * Copyright (C) 2019, STMicroelectronics - All Rights Reserved
> >> + *
> >> + * Driver for STMicroelectronics Serial peripheral interface (SPI)
> >> + */
> >> +
> >
> > ...
> >
> >> +/* STM32_SPI_CR2 bit fields */
> >> +#define SPI_CR2_TSIZE GENMASK(15, 0)
> >
> > So SPI_CR2_TSIZE is (64K - 1)
> >
> > ...
> >
> >> +
> >> +static int stm32_spi_transfer_one(struct stm32_spi_priv *priv,
> >> + struct spi_transfer *t)
> >> +{
> >> + struct device_d *dev = priv->master.dev;
> >> + u32 sr;
> >> + u32 ifcr = 0;
> >> + u32 mode;
> >> + int xfer_status = 0;
> >> +
> >> + if (t->len <= SPI_CR2_TSIZE)
> >> + writel(t->len, priv->base + STM32_SPI_CR2);
> >> + else
> >> + return -EMSGSIZE;
> >
> > So stm32_spi_transfer_one() can transfer no more than (64K - 1).
> > At the other hand imd tends to read more than (64K - 1) from SPI flash, e.g.
> >
> > barebox:/ imd /dev/m25p0
> > imd: error 90
> >
> > Here is my solution for handling SPI flash by stm32 SPI driver (not tested on stm32):
>
> Thanks for the input. I believe I have STM32 with a SPI flash, where
> I should be able to test this. I'll report back once I come around to it.
>
> Can I have your Signed-off-by for this patch here?
Yes, please add my Signed-off-by for the patch.
--
Best regards,
Antony Pavlov
_______________________________________________
barebox mailing list
barebox@xxxxxxxxxxxxxxxxxxx
http://lists.infradead.org/mailman/listinfo/barebox
