LoopVectorization.jl
LoopVectorization.jl copied to clipboard
for >3 loops in a group, don't permute all possible orders
This didn't appear to help compile time performance in the case I was interested in.
Codecov Report
Merging #398 (dc53c94) into main (7a0ffd4) will increase coverage by
0.39%. The diff coverage is100.00%.
:exclamation: Current head dc53c94 differs from pull request most recent head 608c66d. Consider uploading reports for the commit 608c66d to get more accurate results
@@ Coverage Diff @@
## main #398 +/- ##
==========================================
+ Coverage 86.95% 87.34% +0.39%
==========================================
Files 37 38 +1
Lines 9305 9545 +240
==========================================
+ Hits 8091 8337 +246
+ Misses 1214 1208 -6
| Impacted Files | Coverage Δ | |
|---|---|---|
| src/modeling/determinestrategy.jl | 95.62% <100.00%> (+0.67%) |
:arrow_up: |
| src/user_api_conveniences.jl | 90.47% <0.00%> (-9.53%) |
:arrow_down: |
| src/codegen/lower_threads.jl | 48.93% <0.00%> (-4.23%) |
:arrow_down: |
| src/simdfunctionals/map.jl | 88.50% <0.00%> (-4.07%) |
:arrow_down: |
| src/simdfunctionals/vmap_grad_rrule.jl | 90.62% <0.00%> (-1.57%) |
:arrow_down: |
| src/codegen/lower_memory_common.jl | 86.64% <0.00%> (-0.37%) |
:arrow_down: |
| src/condense_loopset.jl | 96.89% <0.00%> (-0.20%) |
:arrow_down: |
| src/codegen/lowering.jl | 92.60% <0.00%> (-0.08%) |
:arrow_down: |
| src/constructors.jl | 98.72% <0.00%> (ø) |
|
| src/getconstindexes.jl | 100.00% <0.00%> (ø) |
|
| ... and 11 more |
Help us with your feedback. Take ten seconds to tell us how you rate us. Have a feature suggestion? Share it here.