[Datasheet] Incorrect constants shown for HASH_DEF, HASH_VALUE, and SIGNATURE block items

[lucasbaizer]

I've found some errors in the datasheet that should be resolved.

First, in 5.9.2.2. "HASH_DEF item", size_lo is shown to be the constant 0x03. This is not correct; the HASH_DEF item takes up two words, not three. When picotool seal is used to sign an image, the HASH_DEF item it generates has size_lo correctly set to 0x02, and that should be reflected in the datasheet.

Second, the "tag types" for HASH_VALUE and SIGNATURE (sections 5.9.2.3 and 5.9.2.4 respectively) are shown to be 0x09 and 0x4b respectively. These are swapped from their actual values. In pico-sdk, these constants are defined as follows:

#define PICOBIN_BLOCK_ITEM_HASH_VALUE                   _u(0x4b)
#define PICOBIN_BLOCK_ITEM_SIGNATURE                    _u(0x09)

Which is the opposite of what's shown in the datasheet.

[lurch]

I've found some errors in the datasheet that should be resolved.

To clarify for others: this is referring to https://datasheets.raspberrypi.com/rp2350/rp2350-datasheet.pdf

First, in 5.9.2.2. "HASH_DEF item", size_lo is shown to be the constant 0x03. This is not correct; the HASH_DEF item takes up two words, not three. When picotool seal is used to sign an image, the HASH_DEF item it generates has size_lo correctly set to 0x02, and that should be reflected in the datasheet.

Yup, sounds like you might have found a typo. pinging @will-v-pi for confirmation.

Second, the "tag types" for HASH_VALUE and SIGNATURE (sections 5.9.2.3 and 5.9.2.4 respectively) are shown to be 0x09 and 0x4b respectively. These are swapped from their actual values.

I believe this is a duplicate of #469 ?

[will-v-pi]

First, in 5.9.2.2. "HASH_DEF item", size_lo is shown to be the constant 0x03. This is not correct; the HASH_DEF item takes up two words, not three. When picotool seal is used to sign an image, the HASH_DEF item it generates has size_lo correctly set to 0x02, and that should be reflected in the datasheet.

Yup, sounds like you might have found a typo. pinging @will-v-pi for confirmation.

Yes, can confirm this is a typo