[RFC PATCH v2 3/4] liveupdate: Replace compatibility string with a 2-bit pair feature header

[Pasha Tatashin]

On 10-26 20:49, David Matlack wrote:
> I still think it's not advisable to define name##_wrap since anyone
> could accidentally start using it on the incoming path and we wouldn't
> notice at first (especially since same kernel Live Update tests would
> keep working totally fine). I also don't think asserting the combined
> size buys us anything. The header size is just a function of the
> number of features, and what we are trying to protect against is
> structs embedded in the payload changing size.

Fair point.

> 
> So I would just do:
> 
>   /* Marks struct name as a KHO serialized payload with a kho_ser_hdr. */
>   #define DECLARE_KHO_SER(name, payload_size) \
>           static_assert(sizeof(struct name) == (payload_size))
> 
>   DECLARE_KHO_SER(luo_memfd_ser, 16);

My main requirement is that DECLARE_KHO_SER() in include/linux/kho/abi/ 
must be enforced by the compiler rather than being just a bare 
static_assert() that someone could forget to add. Since kho/abi/ headers 
contain both versioned structs and unversioned sub-structs, it must be 
mandatory to declare versioned structs in kho/abi/ with their 
static_assert size check before the kho_ser_*() helpers can be used on 
them.

So, how about adding enum names like this, that are going to be required 
in accessor functions later:

#define DECLARE_KHO_SER(name, payload_size, nr_feats)             \
    static_assert(sizeof(struct name) == (payload_size),          \
                  "struct " #name " size mismatch");              \
    enum {                                                        \
        __KHO_SER_SIZE_##name = (payload_size),                   \
        __KHO_SER_NR_FEATS_##name = (nr_feats),                   \
    }

> Plus helpers that allocate and initialize the header and return the
> payload as a typed pointer. Both the outgoing and incoming paths find
> the payload the same way, by computing its offset from the header,
> rather than through a struct with a fixed payload offset:
> 
>   /* outgoing */
>   hdr = kho_ser_alloc(luo_memfd_ser, LUO_MEMFD_NR_FEATS);
>   ser = kho_ser_payload(hdr, luo_memfd_ser);
> 
>   /* incoming */
>   hdr = phys_to_virt(args->serialized_data);
>   ser = kho_ser_payload(hdr, luo_memfd_ser);

These would become, where without enum, the init would fail:

#define kho_ser_init(buf, name)                                     \
      __kho_ser_init(buf, __KHO_SER_SIZE_##name,                    \
                          __KHO_SER_NR_FEATS_##name)

And payload can also check the existance of DECLARE_KHO_SER, using 
(expr1, expr2):

#define kho_ser_payload(hdr, name)                                  \
      ((void)__KHO_SER_SIZE_##name,                                 \
       (struct name *)__kho_ser_payload(hdr))

We can also add an optional kho_ser_alloc(name) to allocate, init, and 
preserve.