yuqei/wycheproof
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Adding timing analysis to the plans for version 0.8.

Browse Source 
NOKEYCHECK=True
PiperOrigin-RevId: 231194032
GitOrigin-RevId: bbe9af9542f60547f7ebbfeb4efe879327e06f50
This commit is contained in:
bleichen
2019-01-28 04:08:10 -08:00
committed by Tink Team
parent ec8cac4845
commit 1025b43dd7
1 changed files with 16 additions and 7 deletions
Download Patch File Download Diff File Expand all files Collapse all files
doc/changes.md
+16 -7
View File
@@ -57,13 +57,22 @@ Additional test vectors\:
### Some potential plans for version 0.8
* Test vectors for long hashes and MACs. I.e. the inputs are strings and
number of repetitions. This allows to add tests for primitives like
HMAC('a' * (2**32+12345)).
* New algorithms: PMAC, X963KDF
* Adding code for capabilities of the providers. Currently this is written in
unstructured form into the logs. Better would be a standalone binary that
writes, JSON or html or text.
* Adding a component for the analysis of key generation and signature
generation. Currently, such tests only run against Java providers. The goal
is to allow any library to be tested.
* Extending tests for timing differences. So far there are only a limited
number of tests against Java providers. The statistical analysis of the
measurments are relatively simple and could probably be improved. One goal
is to separate measuring the timings and doing the statistical analysis by
defining a JSON structure for storing the measurements.
* Adding test vectors for long hashes and MACs. I.e. the inputs are strings
and number of repetitions. This allows to add tests for primitives like
HMAC('a' * (2**32+12345)).
* New algorithms: PMAC, X963KDF
### Some potential plans for later versions
* Adding code for capabilities of the providers. Currently this is written in
unstructured form into the logs. Better would be a standalone binary that
writes, JSON or html or text.
* alternative protocols: PKCS #11, AMD SEV, ...