[dt,otp] Rename misleadingly name field and its documentation #28795
Conversation
pamaury
requested review from
a team,
Razer6,
sameo and
timothytrippel
and removed request for
a team
pamaury
force-pushed
the
dt_otp_rename_field
branch
from
b68a064 to
2ca7d07
Compare
The documentation was probably copy-paste: the digest is not actually an OTP address but a CSR. Rename the field to make it clearer. Signed-off-by: Amaury Pouly <[email protected]>
pamaury
force-pushed
the
dt_otp_rename_field
branch
from
2ca7d07 to
ac1955c
Compare
| name = self.OTP_PARTITION_INFO_STRUCT_DIGEST_ADDR_FIELD_NAME, | ||
| field_type = ScalarType("uint32_t"), | ||
| docstring = "The absolute OTP address at which this partition's digest starts", | ||
| docstring = "The OTP digest CSR (where the digest is buffered) offset for " + |
There was a problem hiding this comment.
Nit: the new wording here suggests that the digest occupies one CSR but in actuality an OTP digest is 8 bytes and falls across 2 CSRs. Maybe this should just be "digest_offset" as a name and the wording should be more like "The OTP address (offset from this partition) at which this partition's digest starts."
There was a problem hiding this comment.
I am not an otp_ctrl expert but I think your phrasing is incorrect because this value is set to the offset of a CSR, e.g. https://opentitan.org/book/hw/top_earlgrey/ip_autogen/otp_ctrl/doc/registers.html#hw_cfg0_digest (you can see the content here).
The idea of using the word "register" is to make it clear that this is the offset of a CSR and not the offset within the OTP partition.
Maybe I could change the wording to "The first (_DIGEST_0) OTP digest CSR" ?
There was a problem hiding this comment.
I see, I misunderstood this as the DAI digest offset rather than the buffered CSR digest (I forgot there was that duplication there and thought that the change was just absolute -> relative address completely 😅) I agree with your new wording.
2 participants
This is a follow-up on #28794 which is a backport of #25975 where the misleading name lead to an error in the ROM. Since master uses the DT, the fix is a bit different but fundamentally achieves the exact same renaming.