[PATCH] hw/ssi: Fix Linux driver init issue with xilinx_spi

[PATCH] hw/ssi: Fix Linux driver init issue with xilinx_spi

[Chris Rauer]

The problem is that the Linux driver expects the master transaction inhibit
bit(R_SPICR_MTI) to be set during driver initialization so that it can
detect the fifo size but QEMU defaults it to zero out of reset.  The
datasheet indicates this bit is active on reset.

See page 25, SPI Control Register section:
https://www.xilinx.com/content/dam/xilinx/support/documents/ip_documentation/axi_quad_spi/v3_2/pg153-axi-quad-spi.pdf

Signed-off-by: Chris Rauer <crauer@google.com>
---
 hw/ssi/xilinx_spi.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/hw/ssi/xilinx_spi.c b/hw/ssi/xilinx_spi.c
index 552927622f..d4de2e7aab 100644
--- a/hw/ssi/xilinx_spi.c
+++ b/hw/ssi/xilinx_spi.c
@@ -156,6 +156,7 @@ static void xlx_spi_do_reset(XilinxSPI *s)
     txfifo_reset(s);
 
     s->regs[R_SPISSR] = ~0;
+    s->regs[R_SPICR] = R_SPICR_MTI;
     xlx_spi_update_irq(s);
     xlx_spi_update_cs(s);
 }
-- 
2.40.0.348.gf938b09366-goog

Re: [PATCH] hw/ssi: Fix Linux driver init issue with xilinx_spi

[Peter Maydell]

Ping to the Xilinx folks, do you want to review this?

thanks
-- PMM

On Thu, 23 Mar 2023 at 18:28, Chris Rauer <crauer@google.com> wrote:
>
> The problem is that the Linux driver expects the master transaction inhibit
> bit(R_SPICR_MTI) to be set during driver initialization so that it can
> detect the fifo size but QEMU defaults it to zero out of reset.  The
> datasheet indicates this bit is active on reset.
>
> See page 25, SPI Control Register section:
> https://www.xilinx.com/content/dam/xilinx/support/documents/ip_documentation/axi_quad_spi/v3_2/pg153-axi-quad-spi.pdf
>
> Signed-off-by: Chris Rauer <crauer@google.com>
> ---
>  hw/ssi/xilinx_spi.c | 1 +
>  1 file changed, 1 insertion(+)
>
> diff --git a/hw/ssi/xilinx_spi.c b/hw/ssi/xilinx_spi.c
> index 552927622f..d4de2e7aab 100644
> --- a/hw/ssi/xilinx_spi.c
> +++ b/hw/ssi/xilinx_spi.c
> @@ -156,6 +156,7 @@ static void xlx_spi_do_reset(XilinxSPI *s)
>      txfifo_reset(s);
>
>      s->regs[R_SPISSR] = ~0;
> +    s->regs[R_SPICR] = R_SPICR_MTI;
>      xlx_spi_update_irq(s);
>      xlx_spi_update_cs(s);
>  }
> --
> 2.40.0.348.gf938b09366-goog
>

Re: [PATCH] hw/ssi: Fix Linux driver init issue with xilinx_spi

[Edgar E. Iglesias]

[-- Attachment #1: Type: text/plain, Size: 1212 bytes --]

On Thu, Mar 23, 2023 at 7:29 PM Chris Rauer <crauer@google.com> wrote:

> The problem is that the Linux driver expects the master transaction inhibit
> bit(R_SPICR_MTI) to be set during driver initialization so that it can
> detect the fifo size but QEMU defaults it to zero out of reset.  The
> datasheet indicates this bit is active on reset.
>
> See page 25, SPI Control Register section:
>
> https://www.xilinx.com/content/dam/xilinx/support/documents/ip_documentation/axi_quad_spi/v3_2/pg153-axi-quad-spi.pdf
>
>
Yes, MTI should be set when the device comes out of reset.

Reviewed-by: Edgar E. Iglesias <edgar@zeroasic.com>



> Signed-off-by: Chris Rauer <crauer@google.com>
> ---
>  hw/ssi/xilinx_spi.c | 1 +
>  1 file changed, 1 insertion(+)
>
> diff --git a/hw/ssi/xilinx_spi.c b/hw/ssi/xilinx_spi.c
> index 552927622f..d4de2e7aab 100644
> --- a/hw/ssi/xilinx_spi.c
> +++ b/hw/ssi/xilinx_spi.c
> @@ -156,6 +156,7 @@ static void xlx_spi_do_reset(XilinxSPI *s)
>      txfifo_reset(s);
>
>      s->regs[R_SPISSR] = ~0;
> +    s->regs[R_SPICR] = R_SPICR_MTI;
>      xlx_spi_update_irq(s);
>      xlx_spi_update_cs(s);
>  }
> --
> 2.40.0.348.gf938b09366-goog
>
>
>

[-- Attachment #2: Type: text/html, Size: 2068 bytes --]

Re: [PATCH] hw/ssi: Fix Linux driver init issue with xilinx_spi

[Peter Maydell]

On Fri, 31 Mar 2023 at 11:09, Edgar E. Iglesias
<edgar.iglesias@gmail.com> wrote:
>
>
> On Thu, Mar 23, 2023 at 7:29 PM Chris Rauer <crauer@google.com> wrote:
>>
>> The problem is that the Linux driver expects the master transaction inhibit
>> bit(R_SPICR_MTI) to be set during driver initialization so that it can
>> detect the fifo size but QEMU defaults it to zero out of reset.  The
>> datasheet indicates this bit is active on reset.
>>
>> See page 25, SPI Control Register section:
>> https://www.xilinx.com/content/dam/xilinx/support/documents/ip_documentation/axi_quad_spi/v3_2/pg153-axi-quad-spi.pdf
>>
>
> Yes, MTI should be set when the device comes out of reset.
>
> Reviewed-by: Edgar E. Iglesias <edgar@zeroasic.com>

Thanks; applied to target-arm.next for 8.0.

-- PMM