CLCLE instruction performance

[JamesWekel]

Fish,

"(or do you prefer to be called Jim?)" Generally Jim, but usually anything written is James.

Here is the updated CLCLE instruction performance pull request with the performance test split into a basic function test and a no-op perfromance test unless manually enabled. I've included the following comment in the CLCLE-04-perfomance.tst for emphasis.

.# ---------------------------------------------------------------------------------- .# This ONLY tests the performance of the CLCLE instruction. .# .# The default is to NOT run performance tests. To enable .# performance test, uncomment the "#r 21fd=ff " .# line below. .# .# Tests: .# 1. CLCLE of 512 bytes .# 2. CLCLE of 512 bytes where operand 1 crosses a page boundary .# 3. CLCLE of 2048 bytes .# 4. CLCLE of 2048 bytes where operand 1 crosses a page boundary .# 5. CLCLE of 2048 bytes where both operand 1 and operand 2 .# crosses a page boundary
.# .# Output: .# For each test, a console line will the generated with timing results, .# as follows: .# / 1,000,000 iterations of CLCLE took 38,698 microseconds
.# / 1,000,000 iterations of CLCLE took 48,617 microseconds
.# / 1,000,000 iterations of CLCLE took 49,178 microseconds
.# / 1,000,000 iterations of CLCLE took 68,355 microseconds
.# / 1,000,000 iterations of CLCLE took 69,991 microseconds .# ----------------------------------------------------------------------------------

The CLCLE-04performance.tst does 5 tests. On my system, the performance results before the change were:
/ 1,000,000 iterations of CLCLE took 9,052,879 microseconds / 1,000,000 iterations of CLCLE took 9,096,818 microseconds / 1,000,000 iterations of CLCLE took 35,928,188 microseconds / 1,000,000 iterations of CLCLE took 35,424,265 microseconds / 1,000,000 iterations of CLCLE took 36,205,255 microseconds

and after the change: / 1,000,000 iterations of CLCLE took 41,159 microseconds / 1,000,000 iterations of CLCLE took 49,770 microseconds / 1,000,000 iterations of CLCLE took 51,996 microseconds / 1,000,000 iterations of CLCLE took 69,810 microseconds / 1,000,000 iterations of CLCLE took 74,077 microseconds

There are 5 tests as I was curious on the impact of operands crossing page boundaries with the revised instruction. But a 99% reduction from the original timing still works for me. And, the reduction is all from your mem_cmp function!

Again, no rush in the review.

Jim

[JamesWekel]

Sorry, cancelled as wrong branch.

Jim