diff options
author | Clifford Wolf <clifford@clifford.at> | 2015-06-06 21:43:33 +0200 |
---|---|---|
committer | Clifford Wolf <clifford@clifford.at> | 2015-06-06 21:43:33 +0200 |
commit | 44ea992fed0d66730d6bdeba740bda53c9e7349f (patch) | |
tree | fa53219ec9af24d4a9e4bdb83937958792999a7f /README.md | |
parent | 90ff3380a4e24762b6dd78ad86fc1beddd44f279 (diff) | |
download | picorv32-44ea992fed0d66730d6bdeba740bda53c9e7349f.tar.gz picorv32-44ea992fed0d66730d6bdeba740bda53c9e7349f.zip |
Updated CPI table in README
Diffstat (limited to 'README.md')
-rw-r--r-- | README.md | 20 |
1 files changed, 11 insertions, 9 deletions
@@ -49,15 +49,17 @@ Performance: The average Cycles per Instruction (CPI) is 5 to 7, depending on the mix of instructions in the code. The CPI for the individual instructions is: -| Instruction | CPI | -| ------------------- | ---:| -| ALU reg + immediate | 4 | -| ALU reg + reg | 5 | -| memory load | 7 | -| memory store | 8 | -| branch, taken | 8 | -| branch, not taken | 5 | -| shift operations | 5+ | +| Instruction | CPI | +| ---------------------| ---:| +| direct jump (jal) | 4 | +| ALU reg + immediate | 4 | +| ALU reg + reg | 5 | +| branch (not taken) | 5 | +| memory load | 7 | +| memory store | 8 | +| branch (taken) | 8 | +| indirect jump (jalr) | 8 | +| shift operations | 5+ | Dhrystone benchmark results: 0.146 DMIPS/MHz (258 Dhrystones/Second/MHz) |