Why the cofactor A from secp256r1 of ECDSA is NULL

[fuxingguo16]

Suggested enhancement

Justification

Mbed TLS needs this because

[tom-cosgrove-arm]

Can you give a bit more information, please? The cofactor for P-256 is 1 - where are you seeing NULL?

[gilles-peskine-arm]

A.p == NULL means that the value of A is equal to P - 3. The implementation of curve arithmetic optimizes this case.

This will be clarified in #6999.

[fuxingguo16]

A.p == NULL means that the value of A is equal to P - 3. The implementation of curve arithmetic optimizes this case.

This will be clarified in #6999.

OK，I got it,Thanks ~~

[fuxingguo16]

In standard Mbed TLS, if the elliptic curve parameters (including the A coefficient) are changed when using a known key pair for signing and verifying, will it have an impact on the results of signing and verifying?

A.p == NULL means that the value of A is equal to P - 3. The implementation of curve arithmetic optimizes this case.

This will be clarified in #6999.

In standard Mbed TLS, if the elliptic curve parameters (including the A coefficient) are changed when using a known key pair for signing and verifying, will it have an impact on the results of signing and verifying?

[gilles-peskine-arm]

Mbed TLS has optimized arithmetic for the curves that it supports, and uses the mbedtls_ecp_group_id to dispatch in some internal functions. We do not support changing the parameters. Only standard curves are supported.