Re: [PATCH 09/15] common: tlv: Add TLV-Signature support

[Ahmad Fatoum <a.fatoum@pengutronix.de>]

Hi Jonas,

On 10/22/25 12:43 PM, Jonas Rebmann wrote:
> Hi Ahmad,
> 
> Just answering to what I don't immediately apply for v2:
> 
> On 2025-10-22 12:00, Ahmad Fatoum wrote:
>>> +/*
>>> + * Retrieve length of header+TLVs (offset of spki hash part of
>>> signature if available)
>>> + */
>>> +
>>> +static inline size_t tlv_spki_hash_offset(const struct tlv_header
>>> *header)
>>> +{
>>> +    size_t ret = size_add(sizeof(struct tlv_header),
>>> get_unaligned_be32(&header->length_tlv));
>>>         return ret; /* SIZE_MAX on overflow */
>>
>> Shouldn't you then check for SIZE_MAX at callsites?
> 
> I chose not to implement this check in all places using those
> size/offset helpers.
> 
> In "[PATCH 01/15] common: clean up TLV code", I check early in the TLV
> handling, that tlv_total_len() doesn't overflow. Later on, it is
> guaranteed that calls to tlv_total_len(), tlv_spki_hash_offset() and the
> such cannot overflow.
> 
> If I where to check at callsites of tlv_spki_hash_offset() I'd need to
> check at all callsites of all TLV size/offset helperss, which seemed
> unnecessary.
> 
> What do you think?

If it's already checked, you don't need to repeat the check.

Thanks,
Ahmad

> 
> Regards,
> Jonas
> 

-- 
Pengutronix e.K.                  |                             |
Steuerwalder Str. 21              | http://www.pengutronix.de/  |
31137 Hildesheim, Germany         | Phone: +49-5121-206917-0    |
Amtsgericht Hildesheim, HRA 2686  | Fax:   +49-5121-206917-5555 |