On 10/12/2015 06:56 PM, Jimmy Zhang wrote:
-----Original Message-----
From: Stephen Warren [mailto:swarren@xxxxxxxxxxxxx]
Sent: Monday, October 12, 2015 3:51 PM
To: Jimmy Zhang
Cc: Allen Martin; Stephen Warren; linux-tegra@xxxxxxxxxxxxxxx
Subject: Re: [cbootimage PATCH v5 2/5] Add support to dump rsa related
fields for t210
On 10/09/2015 07:46 PM, Jimmy Zhang wrote:
Add support to dump rsa pubkey, bct's rsa-pss signature and
bootloader's rsa-pss signature.
diff --git a/src/bct_dump.c b/src/bct_dump.c
+#define ARSE_RSA_PARAM_MAX_BYTES 256
typedef union {
u_int32_t val;
u_int8_t uid[16];
+ u_int8_t rsa_param[ARSE_RSA_PARAM_MAX_BYTES];
} param_types;
Shouldn't that be replaced by something that uses the new get_size()
functionality now implemented in patch 1?
For this data structure, I guess we better stay with a MAX constant.
For display functions, ie
values[i].format(...)
bl_values[j].format(...)
There is no id token being passed in. To use get_size(), all format_xxx function prototype need to be redefined (by adding in id token).
I can submit a new patch if you agree my observations.
If we can't pass the size through all the way to make everything fully
generic, I'd recommend:
a)
Name the field and MAX_BYES constant something more generic so it could
be re-used for arbitrary fields, e.g. such as:
#define PARAM_TYPE_BINARY_DATA_MAX_SIZE 256
u_int8_t binary[PARAM_TYPE_BINARY_DATA_MAX_SIZE]
b)
When filling in that field, call get_size() at that point in time, and
validate that the value is equal to sizeof(binary) or
PARAM_TYPE_BINARY_DATA_MAX_SIZE. At least that way we'll have a check
that the sizes do actually match.
Or, perhaps we can make the field:
struct {
size_t len;
u_int8_t data[PARAM_TYPE_BINARY_DATA_MAX_SIZE];
} binary;
... and hence pass the size around that way?
--
To unsubscribe from this list: send the line "unsubscribe linux-tegra" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
