Hi Szymon,
On Wed, Nov 06, 2013, Szymon Janc wrote:
> +#define check_no_p(opcode) \
> + case opcode: \
> + return len == 0;
> +
> +#define check_fix_p(opcode, msg) \
> + case opcode: \
> + return len == sizeof(struct msg);
> +
> +#define check_var_p(opcode, msg, mp_len_name) \
> + case opcode: \
> + { \
> + const struct msg *mp = (const struct msg *) payload; \
> + return len == (sizeof(*mp) + mp->mp_len_name); \
> + }
> +
> +#define check_props_p(opcode, msg) \
> + case opcode: \
> + { \
> + const struct msg *mp = (const struct msg *) payload; \
> + const struct hal_property *hp = mp->props; \
> + const void *p = mp->props; \
> + int i; \
> + for (i = 0; i < mp->num_props; i++) { \
> + if ((void *) (p + sizeof(*hp) + hp->len) > \
> + (void *) (payload + len)) \
> + return false; \
> + p += sizeof(*hp) + hp->len; \
> + hp = p; \
> + } \
> + return p == payload + len; \
> + }
> +
> +static bool check_cmd_core(uint8_t opcode, const void *payload,
> + uint16_t len)
> +{
> + switch (opcode) {
> + check_fix_p(HAL_OP_REGISTER_MODULE, hal_cmd_register_module);
> + check_fix_p(HAL_OP_UNREGISTER_MODULE, hal_cmd_unregister_module);
> + default:
> + return false;
> + }
> +}
I have a feeling that this kind of approach will eventually not be very
readable/maintainable. Have you looked at how this kind of stuff is
generalized using a simple table, e.g. in btmon or the kernel mgmt
command parsing code?
Johan
--
To unsubscribe from this list: send the line "unsubscribe linux-bluetooth" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
