-
Notifications
You must be signed in to change notification settings - Fork 822
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Feature Request: Comprehensive Espressif Math Tests #6385
Labels
Comments
This was referenced May 5, 2023
gojimmypi
changed the title
Feature Request: Comprehensive Math Tests
Feature Request: Comprehensive Espressif Math Tests
May 5, 2023
I have some additional hardware-specific tests in my test.c that I'll add at some later date. Closing this lingering stale issue for now. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Version
latest master
Description
As noted in #6380, there were some unexpected and incorrect Espressif math results as related to
fp_int
data provided from HW acceleration and the use (or lack thereof) fully initializedfp_int
data structures.There's also the handling of the
used
size value for values such asfp_int = 1
infp_cmp
. (e.g. should values ofused=1
andused=8
be considered equal? currently they are not)As noted in the comment, there may be peripheral consequences to the changes I've proposed in #6382.
I will be looking into any undesired side effects. Indeed if other functions have a similar, incorrect handling of the
used
scope, I could imagine a situation where a test might fail. The test itself may need to be addressed.In any case, this issue is a Feature Request to add more comprehensive math testing to wolfcrypt/test that would have caught the
fp_add
andfp_cmp
problems for Espressif and any other Hardware Acceleration that may be inserted into known, good math libraries.The text was updated successfully, but these errors were encountered: