[PATCH] aspeed: Add support for IBM Bonnell

Guenter Roeck posted 1 patch 2 months, 2 weeks ago
hw/arm/aspeed.c | 87 +++++++++++++++++++++++++++++++++++++++++++++++++
1 file changed, 87 insertions(+)
[PATCH] aspeed: Add support for IBM Bonnell
Posted by Guenter Roeck 2 months, 2 weeks ago
Introduce support for the IBM Bonnell BMC.

Use Rainier machine information for HW strapping and other machine details
since the actual hardware configuration is unknown. I2C device
instantiation is based on the devicetree file in the upstream Linux kernel.

Major difference to Rainier is that the Bonnell devicetree file
instantiates a TPM. It is therefore possible to test TPM functionality
without having to instantiate the TPM manually from the Linux command
line.

Signed-off-by: Guenter Roeck <linux@roeck-us.net>
---
 hw/arm/aspeed.c | 87 +++++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 87 insertions(+)

diff --git a/hw/arm/aspeed.c b/hw/arm/aspeed.c
index fd5603f7aa..4f833c5708 100644
--- a/hw/arm/aspeed.c
+++ b/hw/arm/aspeed.c
@@ -865,6 +865,70 @@ static void rainier_bmc_i2c_init(AspeedMachineState *bmc)
     create_pca9552(soc, 15, 0x60);
 }
 
+static void bonnell_bmc_i2c_init(AspeedMachineState *bmc)
+{
+    AspeedSoCState *soc = bmc->soc;
+    I2CBus *bus;
+
+    bus = aspeed_i2c_get_bus(&soc->i2c, 0);
+
+    at24c_eeprom_init(bus, 0x51, 8 * KiB);      /* atmel,24c64 */
+    /* tca9554@11:20 */
+    i2c_slave_create_simple(bus, "pca9554", 0x20);
+
+    /* ucd90160@2:64 */
+
+    /* fsg032@3:5a */
+    /* fsg032@3:5b */
+
+    bus = aspeed_i2c_get_bus(&soc->i2c, 7);
+
+    /* si7020@7:40 */
+
+    /* Bonnell expects a TMP275 but a TMP105 is compatible */
+    i2c_slave_create_simple(bus, TYPE_TMP105, 0x48);
+    at24c_eeprom_init(bus, 0x50, 8 * KiB);      /* atmel,24c64 */
+    at24c_eeprom_init(bus, 0x51, 8 * KiB);      /* atmel,24c64 */
+    i2c_slave_create_simple(bus, "max31785", 0x52);
+
+    /* pca9551; assume/hope pca9552 is compatible enough */
+    create_pca9552(soc, 7, 0x60);
+
+    /* ibm,op-panel@7:62 */
+    /* dps310@7:76 */
+
+    bus = aspeed_i2c_get_bus(&soc->i2c, 8);
+
+    /* rx8900@8:32 */
+
+    /* Bonnell expects a TMP275 but a TMP105 is compatible */
+    i2c_slave_create_simple(bus, TYPE_TMP105, 0x48);
+    at24c_eeprom_init(bus, 0x51, 16 * KiB);      /* atmel,24c128 */
+
+    /* pca9551@8:60 */
+    create_pca9552(soc, 8, 0x60);
+
+    i2c_slave_create_simple(aspeed_i2c_get_bus(&soc->i2c, 9), "tmp423", 0x4c);
+
+    bus = aspeed_i2c_get_bus(&soc->i2c, 11);
+
+    /* tca9554@11:20 */
+    i2c_slave_create_simple(bus, "pca9554", 0x20);
+    i2c_slave_create_simple(bus, "tmp423", 0x4c);
+    /* pca9849@11:75 */
+    i2c_slave_create_simple(bus, "pca9546", 0x75);
+
+    /* npct75x@12:2e (tpm) */
+
+    /* atmel,24c64 */
+    at24c_eeprom_init(aspeed_i2c_get_bus(&soc->i2c, 12), 0x50, 8 * KiB);
+
+    /* atmel,24c64 */
+    at24c_eeprom_init(aspeed_i2c_get_bus(&soc->i2c, 13), 0x50, 8 * KiB);
+    /* pca9551@13:60 */
+    create_pca9552(soc, 13, 0x60);
+}
+
 static void get_pca9548_channels(I2CBus *bus, uint8_t mux_addr,
                                  I2CBus **channels)
 {
@@ -1488,6 +1552,25 @@ static void aspeed_machine_rainier_class_init(ObjectClass *oc, void *data)
     aspeed_machine_ast2600_class_emmc_init(oc);
 };
 
+static void aspeed_machine_bonnell_class_init(ObjectClass *oc, void *data)
+{
+    MachineClass *mc = MACHINE_CLASS(oc);
+    AspeedMachineClass *amc = ASPEED_MACHINE_CLASS(oc);
+
+    mc->desc       = "IBM Bonnell BMC (Cortex-A7)";
+    amc->soc_name  = "ast2600-a3";
+    amc->hw_strap1 = RAINIER_BMC_HW_STRAP1;
+    amc->hw_strap2 = RAINIER_BMC_HW_STRAP2;
+    amc->fmc_model = "mx66l1g45g";
+    amc->spi_model = "mx66l1g45g";
+    amc->num_cs    = 2;
+    amc->macs_mask  = ASPEED_MAC2_ON | ASPEED_MAC3_ON;
+    amc->i2c_init  = bonnell_bmc_i2c_init;
+    mc->default_ram_size = 1 * GiB;
+    aspeed_machine_class_init_cpus_defaults(mc);
+    aspeed_machine_ast2600_class_emmc_init(oc);
+};
+
 #define FUJI_BMC_RAM_SIZE ASPEED_RAM_SIZE(2 * GiB)
 
 static void aspeed_machine_fuji_class_init(ObjectClass *oc, void *data)
@@ -1776,6 +1859,10 @@ static const TypeInfo aspeed_machine_types[] = {
         .name          = MACHINE_TYPE_NAME("rainier-bmc"),
         .parent        = TYPE_ASPEED_MACHINE,
         .class_init    = aspeed_machine_rainier_class_init,
+    }, {
+        .name          = MACHINE_TYPE_NAME("bonnell-bmc"),
+        .parent        = TYPE_ASPEED_MACHINE,
+        .class_init    = aspeed_machine_bonnell_class_init,
     }, {
         .name          = MACHINE_TYPE_NAME("fuji-bmc"),
         .parent        = TYPE_ASPEED_MACHINE,
-- 
2.45.2
Re: [PATCH] aspeed: Add support for IBM Bonnell
Posted by Cédric Le Goater 2 months, 2 weeks ago
Joel,

Would have access to a Bonnell BMC ? To get the HW strapping
register values.

On 9/6/24 14:35, Guenter Roeck wrote:
> Introduce support for the IBM Bonnell BMC.
> 
> Use Rainier machine information for HW strapping and other machine details
> since the actual hardware configuration is unknown. I2C device
> instantiation is based on the devicetree file in the upstream Linux kernel.
> 
> Major difference to Rainier is that the Bonnell devicetree file
> instantiates a TPM. It is therefore possible to test TPM functionality
> without having to instantiate the TPM manually from the Linux command
> line.
> 
> Signed-off-by: Guenter Roeck <linux@roeck-us.net>
> ---
>   hw/arm/aspeed.c | 87 +++++++++++++++++++++++++++++++++++++++++++++++++
>   1 file changed, 87 insertions(+)
> 
> diff --git a/hw/arm/aspeed.c b/hw/arm/aspeed.c
> index fd5603f7aa..4f833c5708 100644
> --- a/hw/arm/aspeed.c
> +++ b/hw/arm/aspeed.c
> @@ -865,6 +865,70 @@ static void rainier_bmc_i2c_init(AspeedMachineState *bmc)
>       create_pca9552(soc, 15, 0x60);
>   }
>   
> +static void bonnell_bmc_i2c_init(AspeedMachineState *bmc)
> +{
> +    AspeedSoCState *soc = bmc->soc;
> +    I2CBus *bus;
> +
> +    bus = aspeed_i2c_get_bus(&soc->i2c, 0);
> +
> +    at24c_eeprom_init(bus, 0x51, 8 * KiB);      /* atmel,24c64 */
> +    /* tca9554@11:20 */
> +    i2c_slave_create_simple(bus, "pca9554", 0x20);


The other machine tend to simply do :

   i2c_slave_create_simple(aspeed_i2c_get_bus(&soc->i2c, 0), "pca9554", 0x20);

But that's fine.

> +
> +    /* ucd90160@2:64 */
> +
> +    /* fsg032@3:5a */
> +    /* fsg032@3:5b */
> +
> +    bus = aspeed_i2c_get_bus(&soc->i2c, 7);
> +
> +    /* si7020@7:40 */
> +
> +    /* Bonnell expects a TMP275 but a TMP105 is compatible */
> +    i2c_slave_create_simple(bus, TYPE_TMP105, 0x48);
> +    at24c_eeprom_init(bus, 0x50, 8 * KiB);      /* atmel,24c64 */
> +    at24c_eeprom_init(bus, 0x51, 8 * KiB);      /* atmel,24c64 */
> +    i2c_slave_create_simple(bus, "max31785", 0x52);
> +
> +    /* pca9551; assume/hope pca9552 is compatible enough */
> +    create_pca9552(soc, 7, 0x60);
> +
> +    /* ibm,op-panel@7:62 */
> +    /* dps310@7:76 */
> +
> +    bus = aspeed_i2c_get_bus(&soc->i2c, 8);
> +
> +    /* rx8900@8:32 */
> +
> +    /* Bonnell expects a TMP275 but a TMP105 is compatible */
> +    i2c_slave_create_simple(bus, TYPE_TMP105, 0x48);
> +    at24c_eeprom_init(bus, 0x51, 16 * KiB);      /* atmel,24c128 */
> +
> +    /* pca9551@8:60 */
> +    create_pca9552(soc, 8, 0x60);
> +
> +    i2c_slave_create_simple(aspeed_i2c_get_bus(&soc->i2c, 9), "tmp423", 0x4c);
> +
> +    bus = aspeed_i2c_get_bus(&soc->i2c, 11);
> +
> +    /* tca9554@11:20 */
> +    i2c_slave_create_simple(bus, "pca9554", 0x20);
> +    i2c_slave_create_simple(bus, "tmp423", 0x4c);
> +    /* pca9849@11:75 */
> +    i2c_slave_create_simple(bus, "pca9546", 0x75);
> +
> +    /* npct75x@12:2e (tpm) */
> +
> +    /* atmel,24c64 */
> +    at24c_eeprom_init(aspeed_i2c_get_bus(&soc->i2c, 12), 0x50, 8 * KiB);
> +
> +    /* atmel,24c64 */
> +    at24c_eeprom_init(aspeed_i2c_get_bus(&soc->i2c, 13), 0x50, 8 * KiB);
> +    /* pca9551@13:60 */
> +    create_pca9552(soc, 13, 0x60);
> +}
> +
>   static void get_pca9548_channels(I2CBus *bus, uint8_t mux_addr,
>                                    I2CBus **channels)
>   {
> @@ -1488,6 +1552,25 @@ static void aspeed_machine_rainier_class_init(ObjectClass *oc, void *data)
>       aspeed_machine_ast2600_class_emmc_init(oc);
>   };
>   
> +static void aspeed_machine_bonnell_class_init(ObjectClass *oc, void *data)
> +{
> +    MachineClass *mc = MACHINE_CLASS(oc);
> +    AspeedMachineClass *amc = ASPEED_MACHINE_CLASS(oc);
> +
> +    mc->desc       = "IBM Bonnell BMC (Cortex-A7)";
> +    amc->soc_name  = "ast2600-a3";
> +    amc->hw_strap1 = RAINIER_BMC_HW_STRAP1;
> +    amc->hw_strap2 = RAINIER_BMC_HW_STRAP2;
> +    amc->fmc_model = "mx66l1g45g";
> +    amc->spi_model = "mx66l1g45g";
> +    amc->num_cs    = 2;
> +    amc->macs_mask  = ASPEED_MAC2_ON | ASPEED_MAC3_ON;
> +    amc->i2c_init  = bonnell_bmc_i2c_init;
> +    mc->default_ram_size = 1 * GiB;
> +    aspeed_machine_class_init_cpus_defaults(mc);
> +    aspeed_machine_ast2600_class_emmc_init(oc);
> +};
> +
>   #define FUJI_BMC_RAM_SIZE ASPEED_RAM_SIZE(2 * GiB)
>   
>   static void aspeed_machine_fuji_class_init(ObjectClass *oc, void *data)
> @@ -1776,6 +1859,10 @@ static const TypeInfo aspeed_machine_types[] = {
>           .name          = MACHINE_TYPE_NAME("rainier-bmc"),
>           .parent        = TYPE_ASPEED_MACHINE,
>           .class_init    = aspeed_machine_rainier_class_init,
> +    }, {
> +        .name          = MACHINE_TYPE_NAME("bonnell-bmc"),
> +        .parent        = TYPE_ASPEED_MACHINE,
> +        .class_init    = aspeed_machine_bonnell_class_init,
>       }, {
>           .name          = MACHINE_TYPE_NAME("fuji-bmc"),
>           .parent        = TYPE_ASPEED_MACHINE,


Reviewed-by: Cédric Le Goater <clg@kaod.org>

Thanks,

C.



Re: [PATCH] aspeed: Add support for IBM Bonnell
Posted by Guenter Roeck 2 months, 2 weeks ago
On 9/6/24 07:49, Cédric Le Goater wrote:
> Joel,
> 
> Would have access to a Bonnell BMC ? To get the HW strapping
> register values.
> 
> On 9/6/24 14:35, Guenter Roeck wrote:
>> Introduce support for the IBM Bonnell BMC.
>>
>> Use Rainier machine information for HW strapping and other machine details
>> since the actual hardware configuration is unknown. I2C device
>> instantiation is based on the devicetree file in the upstream Linux kernel.
>>
>> Major difference to Rainier is that the Bonnell devicetree file
>> instantiates a TPM. It is therefore possible to test TPM functionality
>> without having to instantiate the TPM manually from the Linux command
>> line.
>>
>> Signed-off-by: Guenter Roeck <linux@roeck-us.net>
>> ---
>>   hw/arm/aspeed.c | 87 +++++++++++++++++++++++++++++++++++++++++++++++++
>>   1 file changed, 87 insertions(+)
>>
>> diff --git a/hw/arm/aspeed.c b/hw/arm/aspeed.c
>> index fd5603f7aa..4f833c5708 100644
>> --- a/hw/arm/aspeed.c
>> +++ b/hw/arm/aspeed.c
>> @@ -865,6 +865,70 @@ static void rainier_bmc_i2c_init(AspeedMachineState *bmc)
>>       create_pca9552(soc, 15, 0x60);
>>   }
>> +static void bonnell_bmc_i2c_init(AspeedMachineState *bmc)
>> +{
>> +    AspeedSoCState *soc = bmc->soc;
>> +    I2CBus *bus;
>> +
>> +    bus = aspeed_i2c_get_bus(&soc->i2c, 0);
>> +
>> +    at24c_eeprom_init(bus, 0x51, 8 * KiB);      /* atmel,24c64 */
>> +    /* tca9554@11:20 */
>> +    i2c_slave_create_simple(bus, "pca9554", 0x20);
> 
> 
> The other machine tend to simply do :
> 
>    i2c_slave_create_simple(aspeed_i2c_get_bus(&soc->i2c, 0), "pca9554", 0x20);
> 
> But that's fine.
> 

I know, I tried to avoid multiple calls to aspeed_i2c_get_bus() if the bus
is needed multiple times.

Thanks,
Guenter