Dmitrii Kovalkov
|
0d1577c5da
|
Better avx2 implementation for rand(). Expected to be ~10% faster
|
2020-05-29 07:35:03 +02:00 |
|
Dmitrii Kovalkov
|
8483dfa272
|
Delete needless rand implementations
|
2020-05-29 07:35:03 +02:00 |
|
Dmitrii Kovalkov
|
d4a8d91789
|
Fix style
|
2020-05-29 07:35:03 +02:00 |
|
Dmitrii Kovalkov
|
feaed1e020
|
rand isn't avx2-vectorizable, I give it up
|
2020-05-29 07:35:02 +02:00 |
|
Dmitrii Kovalkov
|
90bc3e6136
|
More rand implementations
|
2020-05-29 07:35:02 +02:00 |
|
Dmitrii Kovalkov
|
66d530e901
|
fast rand for performance tests
|
2020-05-29 07:35:02 +02:00 |
|
Dmitrii Kovalkov
|
c524642d24
|
Delete awful template PerformanceAdaptro and add simple ImplementationSelector instead
|
2020-05-29 07:35:02 +02:00 |
|
Dmitrii Kovalkov
|
b2b3ba5942
|
Finaly fix clang build
|
2020-05-29 07:35:01 +02:00 |
|
Dmitrii Kovalkov
|
35e4f43ac7
|
Add alternative implenetation for rand
|
2020-05-29 07:35:01 +02:00 |
|
Dmitrii Kovalkov
|
9387981abe
|
Add afwul option for choosing implementation
|
2020-05-29 07:35:01 +02:00 |
|
Dmitrii Kovalkov
|
234a828dd0
|
Add option to disable multitarget build
|
2020-05-29 07:34:11 +02:00 |
|
Dmitrii Kovalkov
|
a4ff8bb933
|
Get rid of handwritten cpu feature detection
|
2020-05-29 07:34:11 +02:00 |
|
Dmitrii Kovalkov
|
f59b13a58d
|
Fix style issues
|
2020-05-29 07:34:10 +02:00 |
|
Dmitrii Kovalkov
|
80ab14e3f9
|
Get rid of copy-paste in PerformanceAdaptor
|
2020-05-29 07:33:38 +02:00 |
|
Dmitrii Kovalkov
|
43f5ca868a
|
Cosmetics
|
2020-05-29 07:33:37 +02:00 |
|
Dmitrii Kovalkov
|
089b3ca008
|
Check target properly
|
2020-05-29 07:33:37 +02:00 |
|
Dmitrii Kovalkov
|
0afa672082
|
make randxorshift great again
|
2020-05-29 07:33:37 +02:00 |
|
Dmitrii Kovalkov
|
808bb14c5c
|
Add xorshift-rand from lemire for comparing performance
|
2020-05-29 07:33:37 +02:00 |
|
Dmitrii Kovalkov
|
acbd3b3a70
|
Compile rand function with AVX
|
2020-05-29 07:31:59 +02:00 |
|
Ivan Lezhankin
|
06446b4f08
|
dbms/ → src/
|
2020-04-03 18:14:31 +03:00 |
|