Closed effectfully closed 1 day ago
/benchmark validation
/benchmark validation
/benchmark validation
Click here to check the status of your benchmark.
Comparing benchmark results of 'validation' on '5eb80f3fe3' (base) and '4a6c11ad07' (PR)
5eb80f3 | 4a6c11a | Change | |
---|---|---|---|
TOTAL | 44.51 ms | 47.06 ms | +5.7% |
Click here to check the status of your benchmark.
WWWWAT hahaha
Comparing benchmark results of 'validation' on '5eb80f3fe3' (base) and '4a6c11ad07' (PR)
5eb80f3 | 4a6c11a | Change | |
---|---|---|---|
TOTAL | 44.32 ms | 47.24 ms | +6.6% |
Click here to check the status of your benchmark.
Comparing benchmark results of 'validation' on '5eb80f3fe3' (base) and '4a6c11ad07' (PR)
5eb80f3 | 4a6c11a | Change | |
---|---|---|---|
TOTAL | 44.59 ms | 47.09 ms | +5.6% |
Those results look pretty conclusive!
/benchmark validation
Click here to check the status of your benchmark.
Comparing benchmark results of 'validation' on '5eb80f3fe' (base) and '4a6c11ad0' (PR)
5eb80f3 | 4a6c11a | Change | |
---|---|---|---|
TOTAL | 44.46 ms | 47.02 ms | +5.8% |
OK, I'm a moron, validation
benchmarks don't include the preparation costs. It's still wild that we get a slowdown here, as if toNamedDeBruijn
was forcing something important that somehow isn't forced by force
. Weird.
/benchmark validation-full
/benchmark validation-full
/benchmark validation-full
Click here to check the status of your benchmark.
Comparing benchmark results of 'validation-full' on '5eb80f3fe' (base) and '4a6c11ad0' (PR)
5eb80f3 | 4a6c11a | Change | |
---|---|---|---|
TOTAL | 111.1 ms | 112.8 ms | +1.6% |
Click here to check the status of your benchmark.
WTF is going on in this PR
Comparing benchmark results of 'validation-full' on '5eb80f3fe' (base) and '4a6c11ad0' (PR)
5eb80f3 | 4a6c11a | Change | |
---|---|---|---|
TOTAL | 111.0 ms | 112.8 ms | +1.6% |
Click here to check the status of your benchmark.
Comparing benchmark results of 'validation-full' on '5eb80f3fe' (base) and '4a6c11ad0' (PR)
5eb80f3 | 4a6c11a | Change | |
---|---|---|---|
TOTAL | 112.8 ms | 111.5 ms | -1.2% |
WTF is going on in this PR
And it is consistent, here's the second output:
and the third one:
It appears that this PR somehow slows down the evaluator, while speeding up preparation, hence why we see results all over the place. How can it possibly slow down the evaluator? I have no idea, but this is likely me screwing something up, because I can't think of a way to make things slower by removing redundant code.
The experiment is over, so I'm closing this PR, but we need to make sure to investigate this weirdness in future.
Let's see how much these nasty name conversions cost us.