Add final single part ops #61
Conversation
Also removed multi-part ops from single part ops table Signed-off-by: Samuel Bailey <[email protected]>
Fixed verify hash typo Signed-off-by: Samuel Bailey <[email protected]>
| @@ -0,0 +1,48 @@ | |||
| # PsaPurgeKey | |||
There was a problem hiding this comment.
Similarly to the comment on your PR on parsec-operations, I think it is best to remove it from now.
| @@ -37,7 +37,7 @@ such signature algorithm: Raw PKCS#1 v1.5 signature. | |||
| **Note:** To perform a hash-and-sign algorithm, the hash must be calculated before passing it to | |||
| this function. This could be done with the operation PsaHashCompute or with a multi-part hash | |||
| operation. Those operations are not yet implemented. Alternatively, to hash and sign a message in a | |||
| single call, you could use PsaSignMessage (not yet implemented). | |||
| single call, you could use PsaSignMessage. | |||
| | Algorithm | Rust client | | ||
| |-----------|-------------| | ||
| | Hmac | ❌ | | ||
| | CbcMac | ❌ | | ||
| | Cmac | ❌ | |
There was a problem hiding this comment.
For things like these algorithms, where the algorithms are supported, but the operations that use them are not, should they be ticks instead?
There was a problem hiding this comment.
I would say no personally. I think this page could be presented in a better way. The first table kind of assume that the operation is fully supported when we only really support one specific algorithm.
Ideally, we would have one massive matrix of combination of providers X combination of operations X combinations of their input parameters.
There was a problem hiding this comment.
And now that all single-part operations are defined, we could think of generating such table programatically, nightly, by either defining new operations that would give us the elvel of support for each operation OR by making test for each one of them. That could be a cool internship project 😉
sbailey-arm
force-pushed
the
add-final-single-part-ops
branch
2 times, most recently
from
cd2d6ed to
20250b6
Compare
|
This should pass CI now the new protobuf contracts are in master. |
There was a problem hiding this comment.
Looks good to me, thank!
The link checker is palying us tricks. I think it's worth investigating why it fails, or using another one. Maybe the GitHub links do some kind of weird redirection that it does not support?
In any case, happy to merge this even if hte CI fails.
| @@ -0,0 +1,36 @@ | |||
| # PsaSignMessage | |||
|
|
|||
| Sign a message with a private key. Opcode: 23 (`0x0017`) | |||
There was a problem hiding this comment.
PsaMacVerify and this one have the same opcode!
src/service_api_coverage.md
Outdated
| | Cfb | ❌ | ❌ | ❌ | | ||
| | Ofb | ❌ | ❌ | ❌ | | ||
| | Xts | ❌ | ❌ | ❌ | | ||
| | EbcNoPadding | ❌ | ❌ | ❌ | |
There was a problem hiding this comment.
| | EbcNoPadding | ❌ | ❌ | ❌ | | |
| | EcbNoPadding | ❌ | ❌ | ❌ | |
There was a problem hiding this comment.
This one and the one below also apply to the client coverage page
src/service_api_coverage.md
Outdated
| | Algorithm | Mbed Crypto provider | PKCS 11 provider | TPM 2.0 provider | | ||
| |--------------|----------------------|------------------|------------------| | ||
| | StreamCipher | ❌ | ❌ | ❌ | | ||
| | Crt | ❌ | ❌ | ❌ | |
There was a problem hiding this comment.
| | Crt | ❌ | ❌ | ❌ | | |
| | Ctr | ❌ | ❌ | ❌ | |
sbailey-arm
force-pushed
the
add-final-single-part-ops
branch
3 times, most recently
from
74778e0 to
39ef2af
Compare
Signed-off-by: Samuel Bailey <[email protected]>
sbailey-arm
force-pushed
the
add-final-single-part-ops
branch
from
39ef2af to
1e1cd61
Compare
Added missing documentation for all single part operations (excluding copy - requires discussion on implementation for Parsec)