Richard Thier
|
22d6631e24
|
added thiersort idea, measure magyar_bucket 1&2
|
2023-04-09 17:20:58 +02:00 |
|
Richard Thier
|
50b1997d5c
|
little optimization to gpts and mine space partition bucket sort
|
2022-12-10 11:04:53 +01:00 |
|
Richard Thier
|
7e21807668
|
added gpt_bucket_sort made with conversing with chatGPT (GPT3 online chat) and added the chat as txt file too with prompts and how it was developed. There were multiple errors, but was interesting
|
2022-12-10 03:13:16 +01:00 |
|
Richard Thier
|
57e328897a
|
added int_to_float.h as might be needed in future for new sort alg idea
|
2022-10-10 19:56:55 +02:00 |
|
Richard Thier
|
d0843cbc40
|
Revert "trying with the restrict keyword - unsuccessful"
This reverts commit 1908687002c085c628853711d26b95e9bc92a1ad.
|
2022-09-01 02:15:43 +02:00 |
|
Richard Thier
|
1908687002
|
trying with the restrict keyword - unsuccessful
|
2022-09-01 02:14:30 +02:00 |
|
Richard Thier
|
5831527f9d
|
removed vectorize makefile part example
|
2022-09-01 01:56:40 +02:00 |
|
Richard Thier
|
7e8aa96a39
|
duplication counting, word-based radix made possible (commented out), modulus impuit, vectorize makefile example
|
2022-09-01 01:56:15 +02:00 |
|
Richard Thier
|
c02aa49f73
|
tweaked to have better MEASURE_ONLY
|
2022-08-31 12:56:51 +02:00 |
|
Richard Thier
|
dac044fe91
|
Revert "trying more ILP in spsort - not much success and will be reverted"
This reverts commit c7e794b7ea1e10f25e64ba29f740890f3aa93b0a.
|
2022-08-16 19:24:35 +02:00 |
|
Richard Thier
|
c7e794b7ea
|
trying more ILP in spsort - not much success and will be reverted
|
2022-08-16 19:24:27 +02:00 |
|
Richard Thier
|
f8793f71ab
|
more ILP in some nearly sorted bucket parts in spsort - rand still not good so much
|
2022-08-16 18:59:18 +02:00 |
|
Richard Thier
|
4bed99751f
|
added feature to use profilers with algorithms more easily
|
2022-08-16 18:37:13 +02:00 |
|
Richard Thier
|
36ea38940b
|
in-place algorithm and fix of perf bug where we not early exited in spsort after insertion sorting small buckets so they were sorted again by us too :-)
|
2022-08-16 18:06:46 +02:00 |
|
Richard Thier
|
b1ab4f2f87
|
tested sp sort and works the first version. Slightly slower on random than std and sometimes slightly faster - but this version do copies that I should optimize out!
|
2022-08-16 16:00:02 +02:00 |
|
Richard Thier
|
edbb59d7a1
|
spsort got twovalue sort special case (no infinite recursion)
|
2022-08-16 15:50:36 +02:00 |
|
Richard Thier
|
6a0a2540bb
|
re-add other test cases - now fails constant because infinite recursion to be fixed for spsort
|
2022-08-16 12:41:19 +02:00 |
|
Richard Thier
|
d0fa5c5b48
|
simplification + fixing right side of array not sorted because -1 counts
|
2022-08-16 12:25:15 +02:00 |
|
Richard Thier
|
6073c03f81
|
temporarily do naive array separate code for debugging
|
2022-08-16 12:16:31 +02:00 |
|
Richard Thier
|
fbea1e607c
|
factored out internal_array_separate(..) to check if it has the errors or not
|
2022-08-16 11:44:34 +02:00 |
|
Richard Thier
|
680936f50a
|
still buggy sp code but differently...
|
2022-08-16 04:13:12 +02:00 |
|
Richard Thier
|
e83392ebaa
|
added "sp" to tests - buggy for now, but at least in const works and inc nearly ok
|
2022-08-16 03:29:08 +02:00 |
|
Richard Thier
|
fad7345a80
|
space partitioning sort first - buggy, but neargood - versions
|
2022-08-16 03:28:06 +02:00 |
|
Richard Thier
|
ee08930cae
|
add -DNDEBUG
|
2022-08-15 23:02:30 +02:00 |
|
Richard Thier
|
ec0f73af01
|
remove session.vim
|
2022-08-15 23:01:55 +02:00 |
|
Richard Thier
|
a2ee3cdb8c
|
make: better makefile
|
2022-01-25 20:04:25 +01:00 |
|
Richard Thier
|
ff05bc2688
|
added if constexpr(..) where it could be
|
2021-12-20 13:29:47 +01:00 |
|
Richard Thier
|
d858f39708
|
Merge branch 'tmp' into ilp-radix-1
|
2021-12-19 22:53:09 +01:00 |
|
Richard Thier
|
c77e592a84
|
mlocks and frewr algorithm both added
|
2021-12-19 21:55:48 +01:00 |
|
|
|
efa2c7bc26
|
old 2007 laptop linux results
|
2021-12-18 20:44:03 +01:00 |
|
|
|
1658e5abbe
|
fine-tuning on linux laptop (just parameters)
|
2021-12-18 20:41:30 +01:00 |
|
Richard Thier
|
a4d50c3309
|
input reduction for testing on less capable machines
|
2021-12-18 19:54:14 +01:00 |
|
Richard Thier
|
da4d122ee1
|
more of latest changes - random weird shiiit
|
2021-12-18 03:49:52 +01:00 |
|
Richard Thier
|
f24b3987c0
|
improved indirections
|
2021-12-18 02:34:22 +01:00 |
|
Richard Thier
|
3b413fcba0
|
removed reference to pointer parameter - a bit better indirections
|
2021-12-18 02:20:42 +01:00 |
|
Richard Thier
|
298edba5d2
|
minor unroll
|
2021-12-18 01:48:42 +01:00 |
|
Richard Thier
|
e7b677e4db
|
basic prefetch optimizations
|
2021-12-18 01:23:06 +01:00 |
|
Richard Thier
|
e5d4ff74ad
|
more manual unrolls
|
2021-12-17 23:37:48 +01:00 |
|
Richard Thier
|
645bc19f19
|
Manual occurence unrolling
|
2021-12-17 22:48:38 +01:00 |
|
Richard Thier
|
be450086b5
|
took out prefetch and added commented out pragmas - they not help
|
2021-12-17 22:09:35 +01:00 |
|
Richard Thier
|
3fdcaad537
|
trying some prefetch - not that good yet
|
2021-12-17 21:42:35 +01:00 |
|
Richard Thier
|
0b4eb5e5a6
|
minor speed tweaks by being able to define the counter type
|
2021-12-17 21:17:53 +01:00 |
|
Richard Thier
|
1686967f10
|
minor tweaks to 4pasu and added 4rot
|
2021-12-17 19:20:58 +01:00 |
|
Richard Thier
|
a878f20100
|
ypsus 4passu method optimized a bit
|
2021-12-15 16:09:40 +01:00 |
|
Richard Thier
|
a947cda58d
|
Revert "vsort version that got slower, but is really funny template code"
This reverts commit fd35dbc51b63fa97ff5a9d7a823cdfa271b99a43.
|
2021-12-15 14:48:27 +01:00 |
|
Richard Thier
|
fd35dbc51b
|
vsort version that got slower, but is really funny template code
|
2021-12-15 14:48:14 +01:00 |
|
Richard Thier
|
bff96c8f7f
|
upgraded vsort a bit (50-100ms)
|
2021-12-15 12:53:00 +01:00 |
|
Richard Thier
|
520db7049d
|
added ypsu-variants of radix-like things
|
2021-12-15 12:52:33 +01:00 |
|
Richard Thier
|
a044787846
|
finally again a real optimization and API for reusal - even faster for non-reused
|
2021-12-15 03:14:35 +01:00 |
|
Richard Thier
|
3490201420
|
further optimization - API change however is not a no-cost abstraction as it makes clang slower than original heap variant and g++ albeit faster than original it does not as fast as hardcoded - will investigave API change
|
2021-12-15 00:43:25 +01:00 |
|
