Re: [PATCHv3 1/2] spi: imx: fix use of native chip-selects with devicetree

[Greg Ungerer <gerg-Td1EMuHUCqxL1ZNQvxDV9g@public.gmane.org>]

Hi Qleksij,

On 24/07/17 16:21, Oleksij Rempel wrote:
> On Thu, Jul 20, 2017 at 11:00:49PM +1000, Greg Ungerer wrote:
>> Hi Oleksij,
>>
>> On 20/07/17 16:34, Oleksij Rempel wrote:
>>> Hi Greg,
>>>
>>> On Tue, Jul 18, 2017 at 09:53:58PM -0300, Fabio Estevam wrote:
>>>> Adding Pengutronix folks on Cc.
>>>>
>>>> On Tue, Jul 11, 2017 at 1:22 AM, Greg Ungerer <gerg-Td1EMuHUCqxL1ZNQvxDV9g@public.gmane.org> wrote:
>>>>> The commonly used mechanism of specifying the hardware or native
>>>>> chip-select on an SPI device in devicetree (that is "cs-gpios = <0>")
>>>>> does not result in the native chip-select being configured for use.
>>>>> So external SPI devices that require use of the native chip-select
>>>>> will not work.
>>>>>
>>>>> You can successfully specify native chip-selects if using a platform
>>>>> setup by specifying the cs-gpio as negative offset by 32. And that
>>>>> works correctly. You cannot use the same method in devicetree.
>>>>>
>>>>> The logic in the spi-imx.c driver during probe uses core spi function
>>>>> of_spi_register_master() in spi.c to parse the "cs-gpios" devicetree tag.
>>>>> For valid GPIO values that will be recorded for use, all other entries in
>>>>> the cs_gpios list will be set to -ENOENT. So entries like "<0>" will be
>>>>> set to -ENOENT in the cs_gpios list.
>>>>>
>>>>> When the SPI device registers are setup the code will use the GPIO
>>>>> listed in the cs_gpios list for the desired chip-select. If the cs_gpio
>>>>> is less then 0 then it is intended to be for a native chip-select, and
>>>>> its cs_gpio value is added to 32 to get the chipselect number to use.
>>>>> Problem is that with devicetree this can only ever be -ENOENT (which
>>>>> is -2), and that alone results in an invalid chip-select number. But also
>>>>> doesn't allow selection of the native chip-select at all.
>>>>>
>>>>> To fix, if the cs_gpio specified for this spi device is not a
>>>>> valid GPIO then use the "chip_select" (that is the native chip-select
>>>>> number) for hardware setup.
>>>>>
>>>>> Signed-off-by: Greg Ungerer <gerg-Td1EMuHUCqxL1ZNQvxDV9g@public.gmane.org>
>>>>> ---
>>>>>    drivers/spi/spi-imx.c | 8 ++++----
>>>>>    1 file changed, 4 insertions(+), 4 deletions(-)
>>>>>
>>>>> diff --git a/drivers/spi/spi-imx.c b/drivers/spi/spi-imx.c
>>>>> index b402530..f4fe66c 100644
>>>>> --- a/drivers/spi/spi-imx.c
>>>>> +++ b/drivers/spi/spi-imx.c
>>>>> @@ -524,8 +524,8 @@ static int mx31_config(struct spi_device *spi, struct spi_imx_config *config)
>>>>>                   reg |= MX31_CSPICTRL_POL;
>>>>>           if (spi->mode & SPI_CS_HIGH)
>>>>>                   reg |= MX31_CSPICTRL_SSPOL;
>>>>> -       if (spi->cs_gpio < 0)
>>>>> -               reg |= (spi->cs_gpio + 32) <<
>>>>> +       if (!gpio_is_valid(spi->cs_gpio))
>>>>> +               reg |= (spi->chip_select) <<
>>>>>                           (is_imx35_cspi(spi_imx) ? MX35_CSPICTRL_CS_SHIFT :
>>>>>                                                     MX31_CSPICTRL_CS_SHIFT);
>>>>>
>>>>> @@ -616,8 +616,8 @@ static int mx21_config(struct spi_device *spi, struct spi_imx_config *config)
>>>>>                   reg |= MX21_CSPICTRL_POL;
>>>>>           if (spi->mode & SPI_CS_HIGH)
>>>>>                   reg |= MX21_CSPICTRL_SSPOL;
>>>>> -       if (spi->cs_gpio < 0)
>>>>> -               reg |= (spi->cs_gpio + 32) << MX21_CSPICTRL_CS_SHIFT;
>>>>> +       if (!gpio_is_valid(spi->cs_gpio))
>>>>> +               reg |= spi->chip_select << MX21_CSPICTRL_CS_SHIFT;
>>>>>
>>>>>           writel(reg, spi_imx->base + MXC_CSPICTRL);
>>>>>
>>>
>>> hm... do I see this correctly, all native chip_selects should
>>> be registered before gpio based CS?
>>
>> I don't follow. The "<0>" must be in the position in the list where
>> you want to use the native chip select. You can't arbitrarily change
>> the order.
>>
>>
>>> For example like this?
>>> cs-gpios = <0>, <&gpio1 1 0>, <&gpio1 2 0>;
>>>
>>> Looks like we don't have any sanity checks for this kind of
>>> configuration:
>>> cs-gpios = <&gpio1 1 0>, <&gpio1 2 0>, <0>;
>>
>> The chip_select is sanity checked in spi_add_device().
>>
>>
>>> We may shift some wired numbers here:
>>> reg |= spi->chip_select << MX21_CSPICTRL_CS_SHIFT;
>>
>> I am not sure I see how that can be the case?
> 
> old and new version of iMX have different amount of native CS.
> I can't find the code which is actually checking if we use right native
> CS-index.
> May be i'm blind :)

I don't think I entirely understand what you are saying. The code at the
top of spi_add_device() [drivers/spi/spi.c] looks like this:

         /* Chipselects are numbered 0..max; validate. */
         if (spi->chip_select >= ctlr->num_chipselect) {
                 dev_err(dev, "cs%d >= max %d\n", spi->chip_select,
                         ctlr->num_chipselect);
                 return -EINVAL;
         }

So it will range check the spi device (spi->chip_select) to be within
the range valid for this SPI controller. That is the very same
spi->chip_select that is used in spi-imx.c to set the register bits
when using a native chip select.

Regards
Greg


--
To unsubscribe from this list: send the line "unsubscribe linux-spi" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html