Re: [PATCH 5/6] aspeed/smc: Let the SSI core layer define the bus name

2022-03-07 Thread Alistair Francis
On Mon, Mar 7, 2022 at 5:34 PM Cédric Le Goater  wrote:
>
> If no id is provided, qdev automatically assigns a unique ename with
> the following pattern ".".
>
> Signed-off-by: Cédric Le Goater 

Reviewed-by: Alistair Francis 

Alistair

> ---
>  hw/ssi/aspeed_smc.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/hw/ssi/aspeed_smc.c b/hw/ssi/aspeed_smc.c
> index f194182beacf..113f31899a6b 100644
> --- a/hw/ssi/aspeed_smc.c
> +++ b/hw/ssi/aspeed_smc.c
> @@ -1130,7 +1130,7 @@ static void aspeed_smc_realize(DeviceState *dev, Error 
> **errp)
>  /* DMA irq. Keep it first for the initialization in the SoC */
>  sysbus_init_irq(sbd, >irq);
>
> -s->spi = ssi_create_bus(dev, "spi");
> +s->spi = ssi_create_bus(dev, NULL);
>
>  /* Setup cs_lines for peripherals */
>  s->cs_lines = g_new0(qemu_irq, asc->max_cs);
> --
> 2.34.1
>
>



Re: [PATCH 5/6] aspeed/smc: Let the SSI core layer define the bus name

2022-03-07 Thread Philippe Mathieu-Daudé

On 7/3/22 08:18, Cédric Le Goater wrote:

If no id is provided, qdev automatically assigns a unique ename with


"an unique name"?


the following pattern ".".


Maybe complete with smth like:

"which avoid bus name collision when using multiple buses."

Reviewed-by: Philippe Mathieu-Daudé 



Signed-off-by: Cédric Le Goater 
---
  hw/ssi/aspeed_smc.c | 2 +-
  1 file changed, 1 insertion(+), 1 deletion(-)




Re: [PATCH 5/6] aspeed/smc: Let the SSI core layer define the bus name

2022-03-07 Thread Cédric Le Goater

On 3/7/22 11:25, Philippe Mathieu-Daudé wrote:

On 7/3/22 08:18, Cédric Le Goater wrote:

If no id is provided, qdev automatically assigns a unique ename with


"an unique name"?


the following pattern ".".


Maybe complete with smth like:

"which avoid bus name collision when using multiple buses."


sure. will do.

The goal behind this patch is to start experimenting with user created
"m25p80" devices such :
 
  -drive file=,format=raw,id=drive0 -device mx66l1g45g,bus=ssi.0,drive=drive0


BMC machines are starting to duplicate and I think, at least for Aspeed,
that we should offer a bare machine per SoC and let the user define
the I2C topology on the command line. The other aspect to cover is
the definition of the flash devices (possibly 3 controllers * 2-3 CS).

Adding a cs= or and addr= is the next step.

Thanks,

C.



[PATCH 5/6] aspeed/smc: Let the SSI core layer define the bus name

2022-03-06 Thread Cédric Le Goater
If no id is provided, qdev automatically assigns a unique ename with
the following pattern ".".

Signed-off-by: Cédric Le Goater 
---
 hw/ssi/aspeed_smc.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/hw/ssi/aspeed_smc.c b/hw/ssi/aspeed_smc.c
index f194182beacf..113f31899a6b 100644
--- a/hw/ssi/aspeed_smc.c
+++ b/hw/ssi/aspeed_smc.c
@@ -1130,7 +1130,7 @@ static void aspeed_smc_realize(DeviceState *dev, Error 
**errp)
 /* DMA irq. Keep it first for the initialization in the SoC */
 sysbus_init_irq(sbd, >irq);
 
-s->spi = ssi_create_bus(dev, "spi");
+s->spi = ssi_create_bus(dev, NULL);
 
 /* Setup cs_lines for peripherals */
 s->cs_lines = g_new0(qemu_irq, asc->max_cs);
-- 
2.34.1