Marshall Lochbaum
|
3aea6aa9d0
|
Boolean prefix sum using AVX2 shuffling instead of pdep
|
2022-04-21 11:43:25 -04:00 |
|
dzaima
|
4c03520a07
|
use maskedLoop in dyarith
|
2022-04-21 18:33:36 +03:00 |
|
dzaima
|
113f72e86d
|
singeli bitarr conversion to wider arrays
|
2022-04-21 18:27:20 +03:00 |
|
dzaima
|
b1fd85621a
|
actually use maskstore
|
2022-04-21 17:48:05 +03:00 |
|
dzaima
|
6f5414b2a5
|
∨`bitarr
|
2022-04-21 15:12:23 +03:00 |
|
dzaima
|
c3504ec42b
|
proper includes
|
2022-04-21 04:13:14 +03:00 |
|
dzaima
|
cd431d3338
|
singeli +`
|
2022-04-21 03:51:37 +03:00 |
|
dzaima
|
11d95b67d5
|
better error messages for no inverse header
|
2022-04-15 23:20:44 +03:00 |
|
dzaima
|
a4a72a78c4
|
singeli select (disabled)
|
2022-04-12 16:45:43 +03:00 |
|
dzaima
|
422dc80283
|
decG round 4
|
2022-04-07 17:19:36 +03:00 |
|
dzaima
|
04dbf9a9a4
|
add another sse3 include
|
2022-04-07 11:40:03 +03:00 |
|
dzaima
|
4a316c9398
|
don't include setjmp.h when CATCH_ERRORS=0
|
2022-04-06 18:52:54 +03:00 |
|
dzaima
|
1609e8b651
|
singeli cleanup
|
2022-04-04 16:28:13 +03:00 |
|
dzaima
|
e031b069fb
|
test special float values
|
2022-04-02 20:00:34 +03:00 |
|
dzaima
|
b329ddee1d
|
unaligned 128-bit vector loads
|
2022-04-02 17:57:49 +03:00 |
|
dzaima
|
40dc184063
|
use singeli for c16lut generation
|
2022-04-02 17:19:20 +03:00 |
|
dzaima
|
49ff5a04ff
|
singeli ≡
|
2022-04-02 16:58:35 +03:00 |
|
dzaima
|
b6dc3cc239
|
/bitarr → i8arr & i16arr
|
2022-04-01 23:30:57 +03:00 |
|
dzaima
|
37bf551e54
|
BMI2 bitarr/i8arr & bitarr/i16arr
|
2022-04-01 22:41:32 +03:00 |
|
dzaima
|
54a9000d88
|
more general any-not-equal check
|
2022-03-28 02:54:13 +03:00 |
|
dzaima
|
09e48854dd
|
skip 16-bit movmsk for addition & subtraction
|
2022-03-28 01:56:46 +03:00 |
|
dzaima
|
60d3ba8c64
|
singeli collect
|
2022-03-20 00:30:06 +02:00 |
|
dzaima
|
4de990a73b
|
bit of singeli cleanup, JIT comment
|
2022-03-18 15:24:13 +02:00 |
|
dzaima
|
47c8c1e5de
|
singeli: iota, better make
|
2022-03-13 18:46:57 +02:00 |
|
dzaima
|
d771f07b30
|
more random singeli additions
|
2022-03-13 15:40:59 +02:00 |
|
dzaima
|
5cd19eafe8
|
vectorized dyadic arith tail element handling
|
2022-02-27 18:46:35 +02:00 |
|
dzaima
|
b1483e14e1
|
random singeli functions
|
2022-02-25 23:46:18 +02:00 |
|
dzaima
|
369831827a
|
better singeli type restriction checks
|
2022-02-25 21:44:38 +02:00 |
|
dzaima
|
7af80e9e3c
|
singeli cast changes
|
2022-02-25 18:22:26 +02:00 |
|
dzaima
|
0d977b73cd
|
clean up some singeli
|
2022-01-25 19:54:18 +02:00 |
|
dzaima
|
3fb793526c
|
)erase; try to not oom on printing errors
|
2022-01-19 01:02:07 +02:00 |
|
dzaima
|
39d1fc0eb3
|
fix precedence
|
2022-01-16 04:50:09 +02:00 |
|
dzaima
|
a6d4c8bb32
|
more singeli definitions
|
2022-01-12 18:29:27 +02:00 |
|
dzaima
|
1770251bad
|
singeli + & - & ×
|
2021-12-18 22:30:00 +02:00 |
|
dzaima
|
e7662b0f6a
|
separate array & scalar cases in arith builtins
|
2021-12-13 00:16:23 +02:00 |
|
dzaima
|
dba145a14b
|
rename folder
|
2021-10-28 17:50:10 +03:00 |
|
dzaima
|
dfa04f4922
|
properly free 𝕩 when needed
|
2021-10-26 16:15:33 +03:00 |
|
dzaima
|
8eb3989d84
|
separate avx file
|
2021-10-25 16:18:03 +03:00 |
|
dzaima
|
071aebd8ea
|
god dammit c compilers, why
|
2021-10-25 15:45:28 +03:00 |
|
dzaima
|
798f6e9e17
|
move length zero check outside singeli
|
2021-10-25 14:00:01 +03:00 |
|
dzaima
|
4e17e28d4a
|
get rid of function pointer cast
|
2021-10-25 13:40:40 +03:00 |
|
dzaima
|
4717023167
|
cmp C cleanup, assert length!=0 before while
|
2021-10-25 12:27:21 +03:00 |
|
dzaima
|
f8798affc4
|
full array-scalar comparison error checking in singeli
|
2021-10-25 01:59:25 +03:00 |
|
dzaima
|
3e372620a2
|
array-scalar comparison function lookup table
|
2021-10-24 21:12:39 +03:00 |
|
dzaima
|
d42dd3dcf9
|
merged typed types
|
2021-10-22 17:11:54 +03:00 |
|
dzaima
|
4ca34e3f8e
|
function lookup table
|
2021-10-22 09:53:40 +03:00 |
|
dzaima
|
5d76148c51
|
lazy singeli bit comparison fns
|
2021-10-22 00:58:34 +03:00 |
|
dzaima
|
71c454235c
|
singeli
|
2021-10-22 00:17:44 +03:00 |
|