Commit Graph

2221 Commits

Author SHA1 Message Date
dzaima
cd7807fba0 NEON constrep 2023-07-10 17:11:15 +03:00
dzaima
1cea42cad3 use getter over 𝕨 for generic 𝕨⍋𝕩
doesn't change perf, but leads to fast amortized long⍋short
2023-07-10 16:50:02 +03:00
dzaima
bf89d2d561 a little size reduction on write_indices 2023-07-10 16:29:21 +03:00
Marshall Lochbaum
f6d1f9fcab Move utilities from bins to appropriate files 2023-07-10 07:14:44 -04:00
Marshall Lochbaum
492e97e2ca Right argument narrowing for numeric Bins 2023-07-10 07:14:44 -04:00
Marshall Lochbaum
9511598aa3 Fast Bins when either argument is boolean 2023-07-10 07:13:21 -04:00
Marshall Lochbaum
1fa00c9c6f Update bins implementation comments 2023-07-10 07:13:21 -04:00
Marshall Lochbaum
46c6d47055 Permutevar instead of shuffle for 4-byte vector binary search 2023-07-10 07:13:21 -04:00
Marshall Lochbaum
fc57e0012d Shuffle-based 4-byte vector binary search 2023-07-10 07:13:21 -04:00
Marshall Lochbaum
af5ba0a2d2 Switch from Eytzinger to linear search on extra lanes 2023-07-10 07:13:21 -04:00
Marshall Lochbaum
64ae8f9afd Slightly better blend pattern for >16-byte bit table 2023-07-10 07:13:21 -04:00
Marshall Lochbaum
c8d20fbf26 And 2-byte vector binary search on 4 lanes 2023-07-10 07:13:21 -04:00
Marshall Lochbaum
fe92f91ca1 2-byte vector binary search on 2 lanes 2023-07-10 07:13:21 -04:00
Marshall Lochbaum
d19df2693a Merge 1-byte table code with normal bins_lookup 2023-07-10 07:13:21 -04:00
Marshall Lochbaum
f36dc7adba Use max-scan instead of plus-scan for 1-byte Bins table 2023-07-10 07:13:21 -04:00
Marshall Lochbaum
8054597004 Handle up to 64 unique values in bit-table 1-byte Bins 2023-07-10 07:13:21 -04:00
Marshall Lochbaum
b62bd02bc0 Precomputed midpoint bits in vector binary search
Clang already optimized to this but it's more explicit
2023-07-10 07:13:21 -04:00
Marshall Lochbaum
7f5ba961c2 Use default bins_lookup if there's no AVX2 2023-07-10 07:13:21 -04:00
Marshall Lochbaum
1080236433 2-byte vector binary searches 2023-07-10 07:13:21 -04:00
Marshall Lochbaum
d665b90bbf Slightly faster binary search pattern with a blend 2023-07-10 07:13:21 -04:00
Marshall Lochbaum
a711eb72eb Table-based 2-byte Bins, using max-scan 2023-07-10 07:13:21 -04:00
Marshall Lochbaum
724f685a57 Cleanup 2023-07-10 07:13:21 -04:00
Marshall Lochbaum
2c9e07f33d 1-byte Bins up to 32 unique eleemnts by unique lookup then index 2023-07-10 07:13:21 -04:00
Marshall Lochbaum
81da9c586f Table-based 1-byte Bins with >1-byte result 2023-07-10 07:13:21 -04:00
Marshall Lochbaum
0346ff7c2d Use type of ≠𝕨 for Singeli Bins result and allocate outside Singeli 2023-07-10 07:13:21 -04:00
Marshall Lochbaum
e0621f99f8 Unroll vector binary searches for different search depths 2023-07-10 07:13:21 -04:00
Marshall Lochbaum
38671e3fe9 Fast handling for length-1 𝕨 in Bins 2023-07-10 07:13:21 -04:00
Marshall Lochbaum
ed9e8b4057 Table-based 1-byte Bins implementations, including AVX2 2023-07-10 07:13:21 -04:00
Marshall Lochbaum
7161689196 Fix bin_search_vec over-writing 2023-07-10 07:13:21 -04:00
Marshall Lochbaum
302d637129 Allocate binary search result in Singeli, using i8 for vectors 2023-07-10 07:13:21 -04:00
Marshall Lochbaum
70249d8b64 In bins, promote to smallest common int type instead of i32 2023-07-10 07:13:21 -04:00
Marshall Lochbaum
4f93b5849b Fix shape of char⍋num result (was deshaped) 2023-07-10 07:13:21 -04:00
Marshall Lochbaum
ec9b875503 Separate is-sorted check from Bins implementation 2023-07-10 07:13:21 -04:00
Marshall Lochbaum
ff9919dde2 AVX2 binary search on one lane of i8 (unused for now) 2023-07-10 07:13:21 -04:00
Marshall Lochbaum
5424c57b7e Use Singeli float binary search 2023-07-10 07:13:21 -04:00
Marshall Lochbaum
ee84f9fe1f 4-way branchless binary search unrolling 2023-07-10 07:13:21 -04:00
Marshall Lochbaum
464a53463d Generic Singeli version of 32-bit binary search 2023-07-10 07:13:21 -04:00
dzaima
6b5815dcc2 fix shape of k/n‿1⥊1 2023-07-10 02:40:36 +03:00
dzaima
8975be082b fix •ns.Get error message 2023-07-09 15:58:31 +03:00
dzaima
5c7934387a update Singeli submodule, match{} → same{} 2023-06-29 15:19:57 +03:00
dzaima
a67a19dd0b move pair{{a,b}} expanding to base.singeli 2023-06-29 01:31:58 +03:00
dzaima
4ee8dfbbda update Singeli submodule, use destructuring parameters 2023-06-28 00:59:28 +03:00
dzaima
3ae9b14d64 separate arith inits, add comments on init order 2023-06-23 12:10:42 +03:00
dzaima
7618dce2f3 fix return value of overflowed dyadic arith 2023-06-23 12:10:42 +03:00
dzaima
5f43bf16b3 add direct typed array type test macro 2023-06-23 12:10:42 +03:00
dzaima
4dbe72ae03 properly indicate estimating Singeli instructions 2023-06-23 11:51:34 +03:00
dzaima
8f8380749b generic Singeli builds 2023-06-16 11:12:15 +03:00
dzaima
5032c9a3ab remove unnecessary 'over _ to' 2023-06-15 19:12:20 +03:00
dzaima
90e7537225 move maskedLoopPositive to @-based 2023-06-15 19:09:53 +03:00
dzaima
dee29a1a02 block → iter 2023-06-15 18:11:57 +03:00
dzaima
9da567cd57 switch to new Singeli block execution 2023-06-15 18:08:57 +03:00
dzaima
2da457dafa fix el_half 2023-06-15 00:32:19 +03:00
dzaima
4260ef93b3 rename away underscores in __adds and __subs
plus add NEON addw, subw, and mulw (which is renamed from mul1)
2023-06-12 00:52:09 +03:00
dzaima
2a52dff482 @-forms of maskedLoop and muLoop 2023-06-12 00:52:09 +03:00
dzaima
fe8892ce6c rearrange shrm/shlm args, slightly better NEON homMask{[16]u8} 2023-06-12 00:52:09 +03:00
dzaima
336f986986 change error messages in ◶ 2023-06-12 00:52:09 +03:00
dzaima
8542ba071d homMaskX, merging multi-arg homMask 2023-06-07 18:14:18 +03:00
dzaima
934a2b5ae3 rearrange things 2023-06-06 21:44:50 +03:00
dzaima
087bf06cac fix type warnings 2023-06-01 20:12:51 +03:00
dzaima
efa1c4a4ff __declspec(dllexport) 2023-06-01 19:04:34 +03:00
dzaima
78bef5fe32 consistent 𝕨⊏𝕩 error messages across argument types 2023-05-30 14:28:36 +03:00
dzaima
2eb12fef82 fix replxx under os=windows 2023-05-30 13:37:15 +03:00
dzaima
7e5b2e0557 don't unroll/vectorize extremely tiny loop 2023-05-30 12:52:17 +03:00
dzaima
5738712bbb improve environ declaration 2023-05-30 12:52:00 +03:00
dzaima
c03dd349e4 disallow •ReBQN constructing namespaces, ignore RETD in interactive REPL 2023-05-27 16:37:42 +03:00
dzaima
da5280021c rename compile → compileAll 2023-05-27 15:38:57 +03:00
dzaima
227dfa20c7 use isatty as a default for notui 2023-05-25 00:16:16 +03:00
dzaima
55e0409582 don't CTZ(0) in gcd 2023-05-24 15:13:25 +03:00
dzaima
c3681a0557 update type number comments 2023-05-22 14:17:53 +03:00
dzaima
3872f5a625 fix 𝕨∾𝕩 reuse logic 2023-05-22 11:40:37 +03:00
dzaima
20a946d085 dynamically switch fork functions 2023-05-22 00:32:44 +03:00
dzaima
b093ebbc66 finish PICK_IDX message 2023-05-22 00:18:59 +03:00
dzaima
f0744f16d0 fix erroring variable read 2023-05-22 00:11:21 +03:00
dzaima
e8caed529c fix JIT SETUi/SETMi/SETCi missing unset variable checks 2023-05-22 00:08:20 +03:00
dzaima
de5bcfe211 static library 2023-05-21 23:47:40 +03:00
dzaima
fd72ea79c8 another gcc warning 2023-05-21 23:39:18 +03:00
dzaima
d15d8febbe slightly less wrong superset fills for 𝕨∾𝕩 2023-05-20 14:28:37 +03:00
dzaima
d6609df820 reserve •CurrentError with namespace 𝕩 2023-05-20 14:28:37 +03:00
dzaima
1e22b48225 in-place ∾ for fillarrs & fillslices 2023-05-20 14:05:22 +03:00
dzaima
bc19e7372d fix pointer check in eachm_fn 2023-05-20 14:05:22 +03:00
dzaima
e781ae21ee give PUSH_COMP access to the updated source string 2023-05-20 13:22:24 +03:00
dzaima
7f46de18fc add missed refcount decrements 2023-05-20 01:30:02 +03:00
dzaima
c9e7dc9976 remove a debug include 2023-05-20 00:01:35 +03:00
dzaima
ad49a9b26d guarantee a space fill to compilation 2023-05-19 22:51:02 +03:00
dzaima
e8d286da3b proper in-place updating in compile() 2023-05-19 22:51:02 +03:00
dzaima
ea70e06eac fix >⟨2,<<3⟩ 2023-05-19 22:51:02 +03:00
dzaima
df348896fd add a NOGC_E for conflicting⌾(list⊸⊏) generic 2023-05-19 22:51:02 +03:00
dzaima
0156d13a5e proper arrow in error messages of ⌾(n⊸↓) vs ⌾(n⊸↑) 2023-05-19 17:32:46 +03:00
dzaima
d930e7f975 add nogc message to thrF 2023-05-19 17:32:14 +03:00
dzaima
a03fab044d •internal.Keep 2023-05-19 15:41:50 +03:00
dzaima
c1488cbf72 faster arr⌾(is⊸⊏)list 2023-05-19 14:15:25 +03:00
dzaima
f6703df0b2 range calculation functions 2023-05-19 14:01:32 +03:00
dzaima
8bfe2b6b5b fix unrolled integer |𝕩 (not used currently) 2023-05-19 13:15:51 +03:00
dzaima
616fd48785 base.singeli additions 2023-05-19 13:15:51 +03:00
dzaima
1ec4415e49 clean up some under functions in sfns.c 2023-05-19 13:15:51 +03:00
dzaima
3977849bc5 fix "_m←¨⋄4 m´⟨⟩" and "4 10´⟨⟩" 2023-05-19 12:58:18 +03:00
Marshall Lochbaum
93e590fd9d One case for dyadic constant-each 2023-05-18 15:40:52 -04:00
Marshall Lochbaum
ed5fac7cc9 Constant and tack cases for Table 2023-05-18 15:11:20 -04:00
Marshall Lochbaum
b5de844736 Constants and tacks with rank, except some dyadic cases 2023-05-18 14:52:13 -04:00
Marshall Lochbaum
7440e2a6bc Constant and trivial cases for fold/insert 2023-05-18 12:11:49 -04:00
Marshall Lochbaum
9dfeb451f8 Handle empty cells in insert_base; fixes bugs in dyadic Insert 2023-05-18 11:53:53 -04:00
Marshall Lochbaum
a8cead1d3d Fast cases for tacks, constant operands, and short arguments in scans 2023-05-18 10:54:56 -04:00
Marshall Lochbaum
1ff8954dd3 Fix leak in dyadic •_timed 2023-05-18 10:54:56 -04:00
Marshall Lochbaum
6ac8944b06 Remove redundant isAtm check in constant case of each_c1 2023-05-18 10:54:56 -04:00
dzaima
b46477450d fix shape of argument passed to F in F⌾(is⊸⊑)
& skip computing the argument if not needed
2023-05-18 16:37:24 +03:00
dzaima
0572fcc5b2 native F⌾(list⊸⊑) & ⌾(listOfLists⊸⊑) 2023-05-18 01:54:01 +03:00
dzaima
d310669ae8 unify F⌾(n⊸⊑) and F⌾⊑ 2023-05-18 00:27:38 +03:00
dzaima
3047ff48b5 Add rank checks to •FChars & •FLines & •FBytes & •Out
& clean up terminology
2023-05-17 21:56:40 +03:00
dzaima
e355395e7a remove duplicated check 2023-05-17 20:53:26 +03:00
dzaima
4f15e1572d disallow •Out ⋈0 2023-05-17 20:50:21 +03:00
dzaima
ec106c0b31 better heuristic for 𝕨∊𝕩 comparison-or path 2023-05-17 18:42:24 +03:00
dzaima
e4d222173d a bunch of fixes for generic architecture & wasm 2023-05-16 00:07:48 +03:00
dzaima
65868a51ba use the type of the correct argument in check 2023-05-13 02:19:10 +03:00
dzaima
f0a9714e8b copy result in tbl_c2 if needed 2023-05-13 00:33:22 +03:00
dzaima
a9478f802f tuning for arithmetic leading axis, F⌜, and reshape-based ⎉ 2023-05-13 00:33:22 +03:00
dzaima
a3e8a1388d move all targets to build.bqn with Singeli & REPLXX enabled by default 2023-05-12 20:51:25 +03:00
dzaima
2f51974c2f use proper pointer conversion macros in ffi.c 2023-05-12 19:36:02 +03:00
dzaima
f229b3c486 fix rtperf
not that anyone uses it anymore
2023-05-12 16:34:23 +03:00
dzaima
a39264533e custom _mm_loadu_si16/_mm_storeu_si16 for old gcc 2023-05-12 16:34:23 +03:00
dzaima
6dc24bd99b free arguments of SIMD 𝕨∊𝕩 & ∊𝕩 2023-05-12 13:51:30 +03:00
Marshall Lochbaum
25883b410c Retune boundary for small-𝕩 linear Member Of 2023-05-11 21:11:43 -04:00
Marshall Lochbaum
53fb8db06f SSSE3 support in bit-table code 2023-05-11 21:05:24 -04:00
Marshall Lochbaum
464dd27a37 SIMD table intialization for Index-of 2023-05-11 21:05:24 -04:00
Marshall Lochbaum
844b537593 Singeli 1-byte table implementations for generic architecture 2023-05-11 21:05:23 -04:00
Marshall Lochbaum
6c81fb3149 Make AVX2 deduplicate branchless 2023-05-11 21:04:58 -04:00
Marshall Lochbaum
1700d84740 AVX2 1-byte Deduplicate implementation 2023-05-11 21:04:58 -04:00
Marshall Lochbaum
9658104f3e Unify AVX2 bit-table code between Member-of and Mark Firsts 2023-05-11 21:04:58 -04:00
Marshall Lochbaum
ba43dae908 Heuristic filtering when AVX2 Mark Firsts sees multiple new values 2023-05-11 21:04:58 -04:00
Marshall Lochbaum
9c7c6aabc5 AVX2 Mark Firsts implementation 2023-05-11 21:04:56 -04:00
Marshall Lochbaum
2fafe11d7c Fix fill_bittab tail masking for t=0 and simplify 2023-05-11 20:59:53 -04:00
Marshall Lochbaum
017e2433fb Fast paths for all empty-argument search cases 2023-05-11 20:59:53 -04:00
Marshall Lochbaum
e45231154c Reverse lookup initialization for 1-byte Member-of 2023-05-11 20:59:53 -04:00
Marshall Lochbaum
0232f91582 AVX2 bit table Member-of 2023-05-11 20:59:52 -04:00
dzaima
03a28e4e34 fix float normalization from a •bit._cast result mutating the original 2023-05-11 22:39:59 +03:00
dzaima
53bacf00c8 native F⟜k⁼𝕩 & k⊸F⁼𝕩 2023-05-11 15:28:40 +03:00
dzaima
08b6ff77c4 c1G & c2G 2023-05-11 15:28:40 +03:00
dzaima
fa4aebbe9c native dyadic inverses of F˜ 2023-05-11 15:28:40 +03:00
dzaima
bae49543be native F˜⁼ 𝕩 2023-05-11 15:28:40 +03:00
dzaima
d1a5ed3e00 err → fatal 2023-05-11 15:28:23 +03:00
dzaima
aa68ed0730 define single ICMP 2023-05-11 15:28:23 +03:00
dzaima
4b0cd3217f proper statement definition of assert 2023-05-11 15:28:23 +03:00
dzaima
fe39fe1932 remove Md1H/MD2R 2023-05-11 15:28:23 +03:00
dzaima
e775202684 describe types in src/README.md 2023-05-10 00:05:29 +03:00
dzaima
a8738c68ed update bytecode submodule 2023-05-09 21:54:01 +03:00
dzaima
5979c0fed9 pass fully native primitives to r1 2023-05-09 21:50:16 +03:00
dzaima
0f2ef4384c don't inline groups_lt 2023-05-09 00:14:14 +03:00
dzaima
300d217ee8 move SIMD constrep functions to a table 2023-05-09 00:14:14 +03:00
dzaima
0598ce8357 use getFillQ less 2023-05-09 00:14:14 +03:00
dzaima
d62c77af41 add threshold for reshape-based ⎉+n path 2023-05-06 16:57:38 +03:00
dzaima
3dcb32a54a fast Arith⎉+n where n is the rank of one argument 2023-05-06 16:28:41 +03:00