Re: [PATCH 6/6] hw/arm/aspeed: Init fby35 BMC FRUID EEPROM

[Peter Delevoryas]

On Mon, Jan 16, 2023 at 01:42:48PM +0100, Cédric Le Goater wrote:
> On 1/14/23 18:01, Peter Delevoryas wrote:
> > Signed-off-by: Peter Delevoryas <peter@pjd.dev>
> > ---
> >   hw/arm/aspeed.c | 49 +++++++++++++++++++++++++++++++++++++++++++++++++
> >   1 file changed, 49 insertions(+)
> > 
> > diff --git a/hw/arm/aspeed.c b/hw/arm/aspeed.c
> > index c929c61d582a..4ac8ff11a835 100644
> > --- a/hw/arm/aspeed.c
> > +++ b/hw/arm/aspeed.c
> > @@ -922,6 +922,52 @@ static void bletchley_bmc_i2c_init(AspeedMachineState 
> > *bmc)
> >       i2c_slave_create_simple(i2c[12], TYPE_PCA9552, 0x67);
> >   }
> > +static const uint8_t fby35_bmc_fruid[] = {
> > +    0x01, 0x00, 0x00, 0x01, 0x0d, 0x00, 0x00, 0xf1, 0x01, 0x0c, 0x00, 0x36,
> > +    0xe6, 0xd0, 0xc6, 0x58, 0x58, 0x58, 0x58, 0x58, 0x58, 0xd2, 0x42, 0x4d,
> > +    0x43, 0x20, 0x53, 0x74, 0x6f, 0x72, 0x61, 0x67, 0x65, 0x20, 0x4d, 0x6f,
> > +    0x64, 0x75, 0x6c, 0x65, 0xcd, 0x58, 0x58, 0x58, 0x58, 0x58, 0x58, 0x58,
> > +    0x58, 0x58, 0x58, 0x58, 0x58, 0x58, 0xce, 0x58, 0x58, 0x58, 0x58, 0x58,
> > +    0x58, 0x58, 0x58, 0x58, 0x58, 0x58, 0x58, 0x58, 0x58, 0xc3, 0x31, 0x2e,
> > +    0x30, 0xc9, 0x58, 0x58, 0x58, 0x58, 0x58, 0x58, 0x58, 0x58, 0x58, 0xd2,
> > +    0x58, 0x58, 0x58, 0x58, 0x58, 0x58, 0x58, 0x58, 0x58, 0x58, 0x58, 0x58,
> > +    0x58, 0x58, 0x58, 0x58, 0x58, 0x58, 0xc1, 0x39, 0x01, 0x0c, 0x00, 0xc6,
> > +    0x58, 0x58, 0x58, 0x58, 0x58, 0x58, 0xd2, 0x59, 0x6f, 0x73, 0x65, 0x6d,
> > +    0x69, 0x74, 0x65, 0x20, 0x56, 0x33, 0x2e, 0x35, 0x20, 0x45, 0x56, 0x54,
> > +    0x32, 0xce, 0x58, 0x58, 0x58, 0x58, 0x58, 0x58, 0x58, 0x58, 0x58, 0x58,
> > +    0x58, 0x58, 0x58, 0x58, 0xc4, 0x45, 0x56, 0x54, 0x32, 0xcd, 0x58, 0x58,
> > +    0x58, 0x58, 0x58, 0x58, 0x58, 0x58, 0x58, 0x58, 0x58, 0x58, 0x58, 0xc7,
> > +    0x58, 0x58, 0x58, 0x58, 0x58, 0x58, 0x58, 0xc3, 0x31, 0x2e, 0x30, 0xc9,
> > +    0x58, 0x58, 0x58, 0x58, 0x58, 0x58, 0x58, 0x58, 0x58, 0xc8, 0x43, 0x6f,
> > +    0x6e, 0x66, 0x69, 0x67, 0x20, 0x41, 0xc1, 0x45, 0xff, 0xff, 0xff, 0xff,
> > +    0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff,
> > +    0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff,
> > +    0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff,
> > +    0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff,
> > +    0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff,
> > +    0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff,
> > +    0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff,
> > +    0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff,
> > +    0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff,
> > +    0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff,
> > +    0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff,
> > +    0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff,
> > +    0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff,
> > +    0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff,
> > +    0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff,
> > +    0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff,
> > +    0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff,
> > +    0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff,
> > +    0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff,
> > +    0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff,
> > +    0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff,
> > +    0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff,
> > +    0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff,
> > +    0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff,
> > +    0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff,
> > +    0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff,
> > +};
> 
> 
> I would introduce a new aspeed_eeprom.c file for these definitions because
> each machine could have its own set of eeproms and aspeed.c is already big
> enough.

+1

> 
> >   static void fby35_i2c_init(AspeedMachineState *bmc)
> >   {
> >       AspeedSoCState *soc = &bmc->soc;
> > @@ -1363,6 +1409,9 @@ static void fby35_reset(MachineState *state, 
> > ShutdownCause reason)
> >       object_property_set_bool(OBJECT(gpio), "gpioB3", false, &error_fatal);
> >       object_property_set_bool(OBJECT(gpio), "gpioB4", false, &error_fatal);
> >       object_property_set_bool(OBJECT(gpio), "gpioB5", false, &error_fatal);
> > +
> > +    at24c_eeprom_write(aspeed_i2c_get_bus(&bmc->soc.i2c, 11),
> > +                       0x54, 0, fby35_bmc_fruid, sizeof(fby35_bmc_fruid));
> >   }
> 
> That's one way to model the default reset values of the eeprom, we would
> loose any writes though.
> 
> I think we should have a reset_data buffer instead, which would be used
> at realize time to set the initial data, if there are no drive backend,
> and at reset if !writable. Something like smbus_eeprom_init_one() does
> without a proper property API.

I actually did it this way downstream[1], but I thought it would be 
controversial
without using properties, and I wasn't sure how to model it with properties.

[1] 
https://github.com/facebook/openbmc/blob/b98f10b7967ebce1f97e14a9a5d4c14f9f7d4c55/common/recipes-devtools/qemu/qemu/0014-hw-nvram-at24c-Add-static-memory-init-option.patch

> 
> We would need some new interface to set a property for a constant buffer
> of uint<>_t values. I don't know how complex that would be. It could be
> useful to other models to define the init state of registers.

Would DEFINE_PROP_ARRAY not work? Won't the properties be constant after
qdev_realize()?

Or if that's not preferable, DEFINE_PROP_LINK (uint8_t *reset_data) +
DEFINE_PROP_UINT32 (reset_data_size)?

Otherwise, if the property stuff is too uncertain, I'd be happy just adding an
adhoc API through the helper like smbus_eeprom_init_one does.

> 
> Thanks,
> 
> C.
> 
> >   static void aspeed_machine_fby35_class_init(ObjectClass *oc, void *data)
>