aboutsummaryrefslogtreecommitdiff
diff options
context:
space:
mode:
authorJoachim StroĢˆmbergson <joachim@secworks.se>2014-02-23 21:19:37 +0100
committerJoachim StroĢˆmbergson <joachim@secworks.se>2014-02-23 21:19:37 +0100
commite24c980e2f06ecdd9be27438f464dc5291d906aa (patch)
treed4cfd2faafdb8dfbc2ccd54ee143686265f64754
parent5e6b7c37996926722cd19038c1813b77cf1302c7 (diff)
Adding more info about the core.
-rw-r--r--README.md16
1 files changed, 14 insertions, 2 deletions
diff --git a/README.md b/README.md
index 36f6d2e..1f04420 100644
--- a/README.md
+++ b/README.md
@@ -2,12 +2,24 @@ sha1
====
## Introduction ##
-Verilog implementation of the SHA-1 cryptgraphic hash function.
-The implementaion follows the specification in NIST FIPS 180-4.
+Verilog implementation of the SHA-1 cryptgraphic hash function. The
+functionality follows the specification in NIST FIPS 180-4.
This core is based on the project at:
https://github.com/secworks/sha1
+The implementation is iterative with one cycle/round. The initialization
+takses one cycle. The W memory is based around a sliding window of 16
+32-bit registers that are updated in sync with the round processing. The
+total latency/message block is 82 cycles.
+
+There is a top level wrapper that provides a 32-bit memory like
+interface for easy integration into a System on Chip (SoC). This
+interface contains mesage block and digest registers to allow a host to
+load the next block while the current block is being processed.
+
+The implementation also includes a functional model written in Python.
+
## Implementation details ##